From cebb216af81d41ccc8e0236c1c2fc71c1cae548f Mon Sep 17 00:00:00 2001 From: wassname Date: Sun, 6 Aug 2017 19:48:33 +1200 Subject: [PATCH] ppo experiments --- pytorch ppo-linear-bn-elu.ipynb | 1025 ++++++++++++++ pytorch ppo-linear.ipynb | 2079 +++++++++++++++++++++++++++++ pytorch ppo-not_shared.ipynb | 2199 +++++++++++++++++++++++++++++++ 3 files changed, 5303 insertions(+) create mode 100644 pytorch ppo-linear-bn-elu.ipynb create mode 100644 pytorch ppo-linear.ipynb create mode 100644 pytorch ppo-not_shared.ipynb diff --git a/pytorch ppo-linear-bn-elu.ipynb b/pytorch ppo-linear-bn-elu.ipynb new file mode 100644 index 0000000..16d6cb3 --- /dev/null +++ b/pytorch ppo-linear-bn-elu.ipynb @@ -0,0 +1,1025 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Pytorch is easier to debug, I like it.\n", + "\n", + "TODO:\n", + "- [x] prioritised experience replay, need to grab loss for each sample, and sample based on loss\n", + "- [ ] check it for my data, can it overfit?, does the normalisation make sense?\n", + "- [x] better metrics\n", + "- [ ] do cnn model\n", + "- [x] read papers\n", + "- [ ] check i'm prioristising by the right things, should lead to lowest loss\n", + "- [ ] test on cartpole\n", + "\n", + "Refs: \n", + "- implementations:\n", + " - PPO\n", + " - **pytorch implementation https://github.com/alexis-jacq/Pytorch-DPPO/blob/master/ppo.py**\n", + " - tensorflow implementation https://github.com/reinforceio/tensorforce/blob/master/tensorforce/models/ppo_model.py\n", + " - Prioritised memory\n", + " - Other\n", + " - http://pytorch.org/tutorials/intermediate/reinforcement_q_learning.html#training\n", + " - https://github.com/pytorch/examples/blob/master/reinforcement_learning/reinforce.py\n", + "- papers:\n", + " - DPPO https://arxiv.org/pdf/1707.02286.pdf\n", + " - PPO \n", + " - https://arxiv.org/abs/1707.06347\n", + " - https://blog.openai.com/openai-baselines-ppo/\n", + " - TRPO https://arxiv.org/abs/1502.05477" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:03.079244Z", + "start_time": "2017-08-06T07:15:02.444122Z" + } + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:__main__:__main__ logger started.\n" + ] + } + ], + "source": [ + "# plotting\n", + "%matplotlib inline\n", + "from matplotlib import pyplot as plt\n", + "import seaborn as sns\n", + "plt.style.use('ggplot')\n", + "\n", + "# numeric\n", + "import numpy as np\n", + "from numpy import random\n", + "import pandas as pd\n", + "\n", + "# utils\n", + "from tqdm import tqdm_notebook as tqdm\n", + "from collections import Counter\n", + "import tempfile\n", + "import logging\n", + "import time\n", + "import datetime\n", + "import random\n", + "\n", + "from collections import OrderedDict\n", + "from IPython.display import display\n", + "from pprint import pprint\n", + "\n", + "# logging\n", + "logger = log = logging.getLogger(__name__)\n", + "log.setLevel(logging.INFO)\n", + "logging.basicConfig()\n", + "log.info('%s logger started.', __name__)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:03.300383Z", + "start_time": "2017-08-06T07:15:03.081000Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "import argparse\n", + "import os\n", + "import sys\n", + "import gym\n", + "from gym import wrappers\n", + "import random\n", + "import numpy as np\n", + "\n", + "import torch\n", + "import torch.optim as optim\n", + "import torch.multiprocessing as mp\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "from torch.autograd import Variable" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:03.325846Z", + "start_time": "2017-08-06T07:15:03.302056Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "import os\n", + "os.sys.path.append(os.path.abspath('.'))\n", + "%reload_ext autoreload\n", + "%autoreload 2" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T06:41:13.661746Z", + "start_time": "2017-08-06T06:41:13.627681Z" + } + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:03.369096Z", + "start_time": "2017-08-06T07:15:03.327470Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "'outputs/agent_portfolio-ddpo/2017-07-21_seperate_weights.pickle'" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "class Params():\n", + " def __init__(self):\n", + " # env\n", + " self.window_length = 50\n", + " # Model\n", + " self.batch_size = 250\n", + " self.lr = 3e-4\n", + " self.gamma = 0.00\n", + " self.gae_param = 0.95\n", + " self.clip = 0.2 # epsilon from eq 7, default 0.2\n", + " self.ent_coeff = 0.\n", + " self.num_epoch = 50\n", + " self.num_steps = 2048*4\n", + " self.time_horizon = 2000000\n", + " self.max_episode_length = 10000\n", + " self.seed = 1\n", + "\n", + "params = Params()\n", + "\n", + "save_path= 'outputs/agent_portfolio-ddpo/{}_seperate_weights.pickle'.format('2017-07-21')\n", + "try:\n", + " os.makedirs(os.path.dirname(save_path))\n", + "except OSError:\n", + " pass\n", + "save_path" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Memory\n", + "refs\n", + "- https://github.com/openai/baselines/blob/master/baselines/deepq/replay_buffer.py\n", + "- https://github.com/jaara/AI-blog/blob/master/Seaquest-DDQN-PER.py" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:03.401745Z", + "start_time": "2017-08-06T07:15:03.370854Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "class ReplayMemory(object):\n", + " def __init__(self, capacity):\n", + " self.capacity = capacity\n", + " self.memory = []\n", + "\n", + " def push(self, events):\n", + " for event in zip(*events):\n", + " self.memory.append(event)\n", + " if len(self.memory)>self.capacity:\n", + " del self.memory[0]\n", + "\n", + " def clear(self):\n", + " self.memory = []\n", + "\n", + " def sample(self, batch_size):\n", + " samples = zip(*random.sample(self.memory, batch_size))\n", + " return map(lambda x: torch.cat(x, 0), samples)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-02T00:55:29.885772Z", + "start_time": "2017-08-02T08:55:29.883459+08:00" + } + }, + "source": [ + "# Enviroment" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:03.659450Z", + "start_time": "2017-08-06T07:15:03.403233Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "(3, 5, 50)" + ] + }, + "execution_count": 6, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from src.environments.portfolio import PortfolioEnv, sharpe, max_drawdown\n", + "\n", + "# we want to pemute the channels a little\n", + "\n", + "class PermutedPortfolioEnv(PortfolioEnv):\n", + " def reset(self, *args, **kwargs):\n", + " return np.transpose(super().reset(*args, **kwargs),(0,1,2))\n", + " def step(self, *args, **kwargs):\n", + " observation, reward, done, info = super().step(*args, **kwargs)\n", + " observation = np.transpose(observation,(2,0,1))\n", + " return observation, reward, done, info\n", + "\n", + "\n", + "df_train = pd.read_hdf('./data/poloniex_30m.hf',key='train')\n", + "env = PermutedPortfolioEnv(\n", + " df=df_train,\n", + " steps=128, \n", + " scale=True, \n", + " augment=0.0025, # let just overfit first,\n", + " trading_cost=0, #0.0025, # let just overfit first,\n", + " window_length = params.window_length, \n", + ")\n", + "env.seed(params.seed)\n", + "env.reset().shape\n", + "\n", + "df_test = pd.read_hdf('./data/poloniex_30m.hf',key='test')\n", + "env_test = PermutedPortfolioEnv(\n", + " df=df_test,\n", + " steps=1280, \n", + " scale=True, \n", + " trading_cost=0, #0.0025, # let just overfit first,\n", + " window_length = params.window_length, \n", + ")\n", + "env_test.seed(params.seed)\n", + "env_test.reset().shape" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Model" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:03.703296Z", + "start_time": "2017-08-06T07:15:03.661087Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "(3, 5, 50)" + ] + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "env.reset().shape\n", + "# 20*50-2" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:03.919796Z", + "start_time": "2017-08-06T07:15:03.705060Z" + } + }, + "outputs": [], + "source": [ + "import torch.nn.init\n", + "\n", + "class GenericSharedModel(nn.Module):\n", + " def __init__(self, inputs, outputs):\n", + " super(GenericSharedModel, self).__init__()\n", + " num_inputs = int(np.prod(env.observation_space.shape))\n", + " num_outputs = int(np.prod(env.action_space.shape))\n", + " \n", + " # hidden layer sizes\n", + " h_size_1 = 100\n", + " h_size_2 = 64\n", + " \n", + " # shared conv block\n", + " self.conv1 = nn.Conv2d(3, 2, (1, 3))\n", + " self.bn_conv1 = nn.BatchNorm2d(2)\n", + " self.conv2 = nn.Conv2d(2, 20, (1, inputs[1] - 2))\n", + " self.bn_conv2 = nn.BatchNorm2d(20)\n", + " \n", + " # Actor mean\n", + " self.fc1 = nn.Linear(20*inputs[0], h_size_1)\n", + " self.bn_fc1 = nn.BatchNorm1d(h_size_1)\n", + " self.fc2 = nn.Linear(h_size_1, h_size_2)\n", + " self.bn_fc2 = nn.BatchNorm1d(h_size_2) \n", + " self.mu = nn.Linear(h_size_2, num_outputs)\n", + " \n", + " # Actor std\n", + " self.log_std = nn.Parameter(torch.zeros(num_outputs))\n", + " \n", + " # Critic\n", + " self.fc1b = nn.Linear(20*inputs[0], h_size_1)\n", + " self.bn_fcb1 = nn.BatchNorm1d(h_size_1)\n", + " self.fc2b = nn.Linear(h_size_1, h_size_2)\n", + " self.bn_fcb2 = nn.BatchNorm1d(h_size_2) \n", + " self.v = nn.Linear(h_size_2,1)\n", + " \n", + " for name, p in self.named_parameters():\n", + " # init parameters like in keras\n", + " if 'bias' in name:\n", + " p.data.fill_(0)\n", + " if ('weight' in name) and ('conv' in name):\n", + " if len(p.size())>1:\n", + " torch.nn.init.xavier_uniform(p)\n", + " else:\n", + " pass # leave as uniform\n", + " \n", + " # mode\n", + " self.train()\n", + "\n", + " def forward(self, inputs):\n", + " # shared conv block\n", + " x = F.elu(self.bn_conv1(self.conv1(inputs)))\n", + " x = F.elu(self.bn_conv2(self.conv2(x)))\n", + " # flatten\n", + " h = x.view(x.size(0),-1)\n", + " \n", + " # the action mean\n", + " x = F.elu(self.bn_fc1(self.fc1(h)))\n", + " x = F.elu(self.bn_fc2(self.fc2(x))) \n", + " mu = self.mu(x)\n", + " \n", + " # the log standard debian of the action\n", + " log_std = torch.exp(self.log_std).unsqueeze(0).expand_as(mu)\n", + " \n", + " # critic\n", + " x = F.elu(self.bn_fcb1(self.fc1b(h)))\n", + " x = F.elu(self.bn_fcb2(self.fc2b(x)))\n", + " v = self.v(x)\n", + " return mu, log_std, v" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T03:41:34.447139Z", + "start_time": "2017-08-06T03:41:34.417249Z" + } + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T03:41:22.398392Z", + "start_time": "2017-08-06T03:41:22.364246Z" + } + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Train" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:03.952805Z", + "start_time": "2017-08-06T07:15:03.921663Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "def mkdir(base, name):\n", + " path = os.path.join(base, name)\n", + " if not os.path.exists(path):\n", + " os.makedirs(path)\n", + " return path" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:03.997538Z", + "start_time": "2017-08-06T07:15:03.954528Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "class Shared_obs_stats():\n", + " \"\"\"Like batchnorm for input data\"\"\"\n", + " def __init__(self, num_inputs):\n", + " self.n = torch.zeros(num_inputs).share_memory_()\n", + " self.mean = torch.zeros(num_inputs).share_memory_()\n", + " self.mean_diff = torch.zeros(num_inputs).share_memory_()\n", + " self.var = torch.zeros(num_inputs).share_memory_()\n", + "\n", + " def observes(self, obs):\n", + " # observation mean var updates\n", + " x = obs.data.squeeze()\n", + " self.n += 1.\n", + " last_mean = self.mean.clone()\n", + " self.mean += (x-self.mean)/self.n\n", + " self.mean_diff += (x-last_mean)*(x-self.mean)\n", + " self.var = torch.clamp(self.mean_diff/self.n, min=1e-2)\n", + "\n", + " def normalize(self, inputs):\n", + " obs_mean = Variable(self.mean.unsqueeze(0).expand_as(inputs))\n", + " obs_std = Variable(torch.sqrt(self.var).unsqueeze(0).expand_as(inputs))\n", + " return torch.clamp((inputs-obs_mean)/obs_std, -5., 5.)" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:04.040170Z", + "start_time": "2017-08-06T07:15:03.999178Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "def normal(x, mu, sigma_sq):\n", + " a = (-1*(x-mu).pow(2)/(2*sigma_sq)).exp()\n", + " b = 1/(2*sigma_sq*np.pi).sqrt()\n", + " return a*b" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:04.078583Z", + "start_time": "2017-08-06T07:15:04.041842Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "GenericSharedModel (\n", + " (conv1): Conv2d(3, 2, kernel_size=(1, 3), stride=(1, 1))\n", + " (bn_conv1): BatchNorm2d(2, eps=1e-05, momentum=0.1, affine=True)\n", + " (conv2): Conv2d(2, 20, kernel_size=(1, 48), stride=(1, 1))\n", + " (bn_conv2): BatchNorm2d(20, eps=1e-05, momentum=0.1, affine=True)\n", + " (fc1): Linear (100 -> 100)\n", + " (bn_fc1): BatchNorm1d(100, eps=1e-05, momentum=0.1, affine=True)\n", + " (fc2): Linear (100 -> 64)\n", + " (bn_fc2): BatchNorm1d(64, eps=1e-05, momentum=0.1, affine=True)\n", + " (mu): Linear (64 -> 6)\n", + " (fc1b): Linear (100 -> 100)\n", + " (bn_fcb1): BatchNorm1d(100, eps=1e-05, momentum=0.1, affine=True)\n", + " (fc2b): Linear (100 -> 64)\n", + " (bn_fcb2): BatchNorm1d(64, eps=1e-05, momentum=0.1, affine=True)\n", + " (v): Linear (64 -> 1)\n", + ")" + ] + }, + "execution_count": 12, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "\n", + "cuda = False\n", + "torch.manual_seed(params.seed)\n", + "work_dir = mkdir('exp', 'ppo')\n", + "monitor_dir = mkdir(work_dir, 'monitor')\n", + "\n", + "# env = gym.make(params.env_name)\n", + "#env = wrappers.Monitor(env, monitor_dir, force=True)\n", + "\n", + "num_inputs = env.observation_space.shape[0]\n", + "num_outputs = env.action_space.shape[0]\n", + "\n", + "\n", + "#initialize network and optimizer\n", + "Model = GenericSharedModel\n", + "model = Model(env.observation_space.shape, env.action_space.shape)\n", + "if cuda: model.cuda()\n", + "\n", + "# shared_obs_stats = Shared_obs_stats(num_inputs)\n", + "optimizer = optim.Adam(model.parameters(), lr=params.lr)\n", + "model" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T07:15:04.138104Z", + "start_time": "2017-08-06T07:15:04.080202Z" + } + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "GenericSharedModel (\n", + " conv1 [ 20]: Conv2d(3, 2, kernel_size=(1, 3), stride=(1, 1))\n", + " bn_conv1 [ 4]: BatchNorm2d(2, eps=1e-05, momentum=0.1, affine=True)\n", + " conv2 [ 1940]: Conv2d(2, 20, kernel_size=(1, 48), stride=(1, 1))\n", + " bn_conv2 [ 40]: BatchNorm2d(20, eps=1e-05, momentum=0.1, affine=True)\n", + " fc1 [ 10100]: Linear (100 -> 100)\n", + " bn_fc1 [ 200]: BatchNorm1d(100, eps=1e-05, momentum=0.1, affine=True)\n", + " fc2 [ 6464]: Linear (100 -> 64)\n", + " bn_fc2 [ 128]: BatchNorm1d(64, eps=1e-05, momentum=0.1, affine=True)\n", + " mu [ 390]: Linear (64 -> 6)\n", + " fc1b [ 10100]: Linear (100 -> 100)\n", + " bn_fcb1 [ 200]: BatchNorm1d(100, eps=1e-05, momentum=0.1, affine=True)\n", + " fc2b [ 6464]: Linear (100 -> 64)\n", + " bn_fcb2 [ 128]: BatchNorm1d(64, eps=1e-05, momentum=0.1, affine=True)\n", + " v [ 65]: Linear (64 -> 1)\n", + ")\n" + ] + } + ], + "source": [ + "from torch.nn.modules.module import _addindent\n", + "import torch\n", + "import numpy as np\n", + "def torch_summarize(model, show_weights=False, show_parameters=True):\n", + " \"\"\"Summarizes torch model by showing trainable parameters and weights\"\"\"\n", + " tmpstr = model.__class__.__name__ + ' (\\n'\n", + " for key, module in model._modules.items():\n", + " # if it contains layers let call it recurvisvly to get params and weights\n", + " if type(module) in [\n", + " torch.nn.modules.container.Container,\n", + " torch.nn.modules.container.Sequential\n", + " ]:\n", + " modstr = torch_summarize(module)\n", + " else:\n", + " modstr = module.__repr__()\n", + " modstr = _addindent(modstr, 2)\n", + " \n", + " params = sum([np.prod(p.size()) for p in module.parameters()])\n", + " weights = tuple([tuple(p.size()) for p in module.parameters()])\n", + " \n", + " tmpstr += ' {:15.15} '.format(key) + '[{: 7.7g}]: '.format(params) + modstr \n", + " if show_weights:\n", + " tmpstr += ', weights={}'.format(weights)\n", + " tmpstr += '\\n' \n", + "\n", + " tmpstr = tmpstr + ')'\n", + " return tmpstr\n", + "\n", + "print(torch_summarize(model))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "start_time": "2017-08-06T07:15:06.348Z" + } + }, + "outputs": [], + "source": [ + "memory = ReplayMemory(params.num_steps)\n", + "# memory = PrioritisedReplayMemory(params.num_steps)\n", + "\n", + "num_inputs = int(np.prod(env.observation_space.shape))\n", + "num_outputs = int(np.prod(env.action_space.shape))\n", + "\n", + "state = env.reset()\n", + "state = Variable(torch.Tensor(state).unsqueeze(0))\n", + "done = True\n", + "episode_length = 0\n", + "reports = []" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "start_time": "2017-08-06T07:15:06.352Z" + }, + "scrolled": true + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Widget Javascript not detected. It may not be installed or enabled properly.\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "108950214b744228a5fb384f1bdddf12" + } + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "episode=63, loss=0.004864, market_value=1.048, cash_bias=0.1248, reward=-3.866e-06, portfolio_value=1.041\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/wassname/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/torch/serialization.py:147: UserWarning: Couldn't retrieve source code for container of type GenericSharedModel. It won't be checked for correctness upon loading.\n", + " \"type \" + obj.__name__ + \". It won't be checked \"\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "episode=127, loss=0.00192, market_value=1.038, cash_bias=0.1695, reward=2.815e-06, portfolio_value=1.041\n", + "16384/|/av_reward= 0.00000326 1%|| 16384/2000000 [02:30<4:03:17, 135.89steps/s]episode=191, loss=0.002165, market_value=1.028, cash_bias=0.2029, reward=7.439e-06, portfolio_value=1.037\n", + "episode=255, loss=0.002224, market_value=1.034, cash_bias=0.2108, reward=4.079e-06, portfolio_value=1.052\n", + "40960/|/av_reward= 0.00057724 2%|| 40960/2000000 [06:20<4:13:53, 128.60steps/s]episode=319, loss=0.002281, market_value=1.042, cash_bias=0.2069, reward=2.316e-06, portfolio_value=1.032\n", + "49152/|/av_reward=-0.00002268 2%|| 49152/2000000 [07:40<3:59:28, 135.77steps/s]episode=383, loss=0.002313, market_value=1.054, cash_bias=0.2293, reward=6.876e-06, portfolio_value=1.047\n", + "episode=447, loss=0.003132, market_value=1.04, cash_bias=0.1413, reward=1.032e-05, portfolio_value=1.027\n", + "65536/|/av_reward=-0.00140124 3%|| 65536/2000000 [10:10<4:00:17, 134.17steps/s]episode=511, loss=0.00504, market_value=1.041, cash_bias=0.165, reward=2.664e-07, portfolio_value=1.015\n", + "episode=575, loss=0.01815, market_value=1.029, cash_bias=0.2184, reward=5.11e-06, portfolio_value=1.027\n", + "81920/|/av_reward= 0.00094801 4%|| 81920/2000000 [12:40<3:56:19, 135.27steps/s]episode=639, loss=0.02324, market_value=1.032, cash_bias=0.1785, reward=-1.139e-05, portfolio_value=1.023\n", + "90112/|/av_reward= 0.00013254 5%|| 90112/2000000 [14:20<5:48:07, 91.44steps/s]episode=703, loss=0.03878, market_value=1.039, cash_bias=0.2411, reward=2.13e-06, portfolio_value=1.044\n", + "98304/|/av_reward= 0.00021603 5%|| 98304/2000000 [16:01<5:02:10, 104.89steps/s]episode=767, loss=0.07314, market_value=1.021, cash_bias=0.1886, reward=1.007e-05, portfolio_value=1.024\n", + "106496/|/av_reward=-0.00005326 5%|| 106496/2000000 [17:41<5:23:23, 97.58steps/s]episode=831, loss=0.1157, market_value=1.022, cash_bias=0.2444, reward=-6.468e-06, portfolio_value=1.033\n", + "114688/|/av_reward=-0.00043878 6%|| 114688/2000000 [19:01<3:56:44, 132.73steps/s]episode=895, loss=0.2048, market_value=1.045, cash_bias=0.2167, reward=1.045e-05, portfolio_value=1.049\n", + "122880/|/av_reward= 0.00089497 6%|| 122880/2000000 [20:21<3:54:27, 133.44steps/s]episode=959, loss=0.3454, market_value=1.06, cash_bias=0.2495, reward=-5.462e-06, portfolio_value=1.063\n", + "episode=1023, loss=0.4507, market_value=1.044, cash_bias=0.2098, reward=3.602e-06, portfolio_value=1.054\n", + "139264/|/av_reward=-0.00036039 7%|| 139264/2000000 [22:51<3:52:04, 133.63steps/s]episode=1087, loss=0.5545, market_value=1.036, cash_bias=0.2722, reward=-3.685e-06, portfolio_value=1.04\n", + "episode=1151, loss=0.5516, market_value=1.029, cash_bias=0.1674, reward=1.566e-07, portfolio_value=1.028\n", + "147456/|/av_reward= 0.00095968 7%|| 147456/2000000 [24:11<3:48:54, 134.88steps/s]episode=1215, loss=0.5942, market_value=1.023, cash_bias=0.1906, reward=-8.688e-06, portfolio_value=1.005\n", + "episode=1279, loss=0.6513, market_value=1.021, cash_bias=0.1861, reward=2.197e-05, portfolio_value=1.032\n", + "episode=1343, loss=0.6651, market_value=1.024, cash_bias=0.1387, reward=-1.033e-06, portfolio_value=1.02\n", + "180224/|/av_reward= 0.00111041 9%|| 180224/2000000 [29:32<3:54:44, 129.20steps/s]episode=1407, loss=0.899, market_value=1.034, cash_bias=0.134, reward=-2.297e-06, portfolio_value=1.04\n", + "188416/|/av_reward= 0.00009518 9%|| 188416/2000000 [30:52<3:43:05, 135.34steps/s]episode=1471, loss=0.8522, market_value=1.03, cash_bias=0.2707, reward=3.024e-06, portfolio_value=1.034\n", + "episode=1535, loss=0.8296, market_value=1.022, cash_bias=0.2185, reward=-1.355e-07, portfolio_value=1.013\n" + ] + } + ], + "source": [ + "\n", + "\n", + "with tqdm(total=params.time_horizon, mininterval=2, unit='steps') as p:\n", + " episode = -1 \n", + " steps = 0\n", + " # horizon loop\n", + " while steps < params.time_horizon:\n", + " infos = []\n", + " episode_length = 0\n", + " # Sample data from the policy\n", + " while (len(memory.memory) < params.num_steps):\n", + " states = []\n", + " actions = []\n", + " rewards = []\n", + " values = []\n", + " returns = []\n", + " advantages = []\n", + " av_reward = 0\n", + " cum_reward = 0\n", + " cum_done = 0\n", + " # n steps loops\n", + " for step in range(params.num_steps):\n", + " # shared_obs_stats.observes(state)\n", + " # state = shared_obs_stats.normalize(state)\n", + " states.append(state)\n", + " \n", + " mu, sigma_sq, v = model(state)\n", + " eps = torch.randn(mu.size())\n", + " action = (mu + sigma_sq.sqrt() * Variable(eps))\n", + " env_action = action.data.squeeze().numpy()\n", + " state, reward, done, info = env.step(env_action)\n", + " done = (done or episode_length >= params.max_episode_length)\n", + " \n", + " cum_reward += reward\n", + " reward = max(min(reward, 1), -1)\n", + " rewards.append(reward)\n", + " actions.append(action)\n", + " values.append(v)\n", + " \n", + " steps+=1 \n", + " p.update(1)\n", + " if done:\n", + " episode += 1\n", + " cum_done += 1\n", + " av_reward += cum_reward\n", + " p.desc='av_reward={: 2.8f}'.format(av_reward / float(cum_done))\n", + " cum_reward = 0\n", + " episode_length = 0\n", + " infos.append(info)\n", + " state = env.reset()\n", + " \n", + " state = Variable(torch.Tensor(state).unsqueeze(0))\n", + " \n", + " if done:\n", + " break\n", + " \n", + " # one last step\n", + " R = torch.zeros(1, 1)\n", + " if not done:\n", + " _, _, v = model(state)\n", + " R = v.data\n", + " \n", + " # compute returns and GAE(lambda) advantages:\n", + " values.append(Variable(R))\n", + " R = Variable(R)\n", + " A = Variable(torch.zeros(1, 1))\n", + " for i in reversed(range(len(rewards))):\n", + " td = rewards[i] + params.gamma*values[i+1].data[0,0] - values[i].data[0,0]\n", + " A = float(td) + params.gamma * params.gae_param * A\n", + " advantages.insert(0, A)\n", + " R = A + values[i]\n", + " returns.insert(0, R)\n", + " \n", + " # store useful info:\n", + " memory.push([states, actions, returns, advantages])\n", + " \n", + "\n", + " # perform several epochs of optimization on the sampled data\n", + " model_old = Model(env.observation_space.shape,\n", + " env.action_space.shape)\n", + " model_old.load_state_dict(model.state_dict())\n", + " if cuda: model_old.cuda()\n", + " av_loss = 0\n", + " for k in range(params.num_epoch):\n", + " # cf https://github.com/openai/baselines/blob/master/baselines/pposgd/pposgd_simple.py\n", + " batch_states, batch_actions, batch_returns, batch_advantages = memory.sample(\n", + " params.batch_size)\n", + " \n", + " # old probas\n", + " mu_old, sigma_sq_old, v_pred_old = model_old(batch_states.detach())\n", + " probs_old = normal(batch_actions, mu_old, sigma_sq_old)\n", + " \n", + " # new probas\n", + " mu, sigma_sq, v_pred = model(batch_states)\n", + " probs = normal(batch_actions, mu, sigma_sq)\n", + " \n", + " # ratio\n", + " ratio = probs / (1e-15 + probs_old)\n", + " \n", + " # surrogate clip loss\n", + " surr1 = ratio * torch.cat([batch_advantages]*num_outputs,1) # surrogate from conservative policy iteration\n", + " surr2 = ratio.clamp(1-params.clip, 1+params.clip) * torch.cat([batch_advantages]*num_outputs,1)\n", + " loss_clip = -torch.mean(torch.min(surr1, surr2))\n", + " # should this be a mean along axis 0?\n", + " \n", + " # state-value function loss, do we even need this if they don't share params?\n", + " vfloss1 = (v_pred - batch_returns)**2\n", + " v_pred_clipped = v_pred_old + (v_pred - v_pred_old).clamp(-params.clip, params.clip)\n", + " vfloss2 = (v_pred_clipped - batch_returns)**2\n", + " loss_value = 0.5 * torch.mean(torch.max(vfloss1, vfloss2))\n", + " # should this be a mean along axis 0?\n", + " \n", + " # loss on entropy bonus to ensure sufficient exploration\n", + " loss_ent = -params.ent_coeff*torch.mean(probs*torch.log(probs+1e-5))\n", + " \n", + " # total\n", + " total_loss = (loss_clip + loss_value + loss_ent)\n", + "# total_loss = (loss_clip - loss_value + loss_ent)\n", + " av_loss += loss_value.data[0] / float(params.num_epoch)\n", + " \n", + " # before step, update old_model:\n", + " model_old.load_state_dict(model.state_dict())\n", + " \n", + " # step\n", + " optimizer.zero_grad()\n", + " total_loss.backward(retain_variables=True)\n", + " optimizer.step()\n", + " \n", + " # t finish, print:\n", + " df_infos = pd.DataFrame(infos)\n", + " \n", + " report=OrderedDict(\n", + " episode=episode,\n", + "# reward=av_reward / float(cum_done),\n", + " loss=av_loss,\n", + " cash_bias=df_infos.cash_bias.mean(),\n", + " market_value=df_infos.market_value.mean(),\n", + " portfolio_value=df_infos.portfolio_value.mean(),\n", + " reward=df_infos.reward.mean()\n", + " )\n", + " \n", + " s = ', '.join(['{}={:2.4g}'.format(key,value) for key,value in report.items()])\n", + " print(s)\n", + " \n", + " reports.append(report)\n", + " \n", + " memory.clear()\n", + " torch.save(model_old, save_path)\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T03:48:54.472250Z", + "start_time": "2017-08-06T03:48:54.442267Z" + } + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "start_time": "2017-08-06T07:15:06.359Z" + } + }, + "outputs": [], + "source": [ + "# show progress\n", + "df=pd.DataFrame(reports)\n", + "g = sns.jointplot(x=\"episode\", y=\"loss\", data=df, kind=\"reg\", size=10)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "start_time": "2017-08-06T07:15:06.364Z" + }, + "scrolled": false + }, + "outputs": [], + "source": [ + "# env_test = env\n", + "\n", + "# Test\n", + "for i in range(10):\n", + " model.train(False)\n", + " state = env_test.reset()\n", + " for i in range(250):\n", + " state = Variable(torch.Tensor(state).unsqueeze(0))\n", + " mu, sigma_sq, v = model(state)\n", + " eps = torch.randn(mu.size())\n", + " action = (mu + sigma_sq.sqrt() * Variable(eps))\n", + " env_action = action.data.squeeze().numpy()\n", + " state, reward, done, info = env_test.step(env_action)\n", + " if done:\n", + " break\n", + "\n", + " env_test.plot()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T03:52:12.779572Z", + "start_time": "2017-08-06T03:52:12.743767Z" + } + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "jupyter3", + "language": "python", + "name": "jupyter3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.5.3" + }, + "toc": { + "colors": { + "hover_highlight": "#DAA520", + "navigate_num": "#000000", + "navigate_text": "#333333", + "running_highlight": "#FF0000", + "selected_highlight": "#FFD700", + "sidebar_border": "#EEEEEE", + "wrapper_background": "#FFFFFF" + }, + "moveMenuLeft": true, + "nav_menu": { + "height": "85px", + "width": "252px" + }, + "navigate_menu": true, + "number_sections": true, + "sideBar": true, + "threshold": 4, + "toc_cell": false, + "toc_section_display": "block", + "toc_window_display": false, + "widenNotebook": false + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/pytorch ppo-linear.ipynb b/pytorch ppo-linear.ipynb new file mode 100644 index 0000000..43e05c3 --- /dev/null +++ b/pytorch ppo-linear.ipynb @@ -0,0 +1,2079 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Pytorch is easier to debug, I like it.\n", + "\n", + "Refs: \n", + "- implementations:\n", + " - PPO\n", + " - **pytorch implementation https://github.com/alexis-jacq/Pytorch-DPPO/blob/master/ppo.py**\n", + " - tensorflow implementation https://github.com/reinforceio/tensorforce/blob/master/tensorforce/models/ppo_model.py\n", + " - Prioritised memory\n", + " - Other\n", + " - http://pytorch.org/tutorials/intermediate/reinforcement_q_learning.html#training\n", + " - https://github.com/pytorch/examples/blob/master/reinforcement_learning/reinforce.py\n", + "- papers:\n", + " - DPPO https://arxiv.org/pdf/1707.02286.pdf\n", + " - PPO \n", + " - https://arxiv.org/abs/1707.06347\n", + " - https://blog.openai.com/openai-baselines-ppo/\n", + " - TRPO https://arxiv.org/abs/1502.05477" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:23.297407Z", + "start_time": "2017-08-05T23:33:22.656714Z" + } + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:__main__:__main__ logger started.\n" + ] + } + ], + "source": [ + "# plotting\n", + "%matplotlib inline\n", + "from matplotlib import pyplot as plt\n", + "import seaborn as sns\n", + "plt.style.use('ggplot')\n", + "\n", + "# numeric\n", + "import numpy as np\n", + "from numpy import random\n", + "import pandas as pd\n", + "\n", + "# utils\n", + "from tqdm import tqdm_notebook as tqdm\n", + "from collections import Counter\n", + "import tempfile\n", + "import logging\n", + "import time\n", + "import datetime\n", + "import random\n", + "\n", + "from collections import OrderedDict\n", + "from IPython.display import display\n", + "from pprint import pprint\n", + "\n", + "# logging\n", + "logger = log = logging.getLogger(__name__)\n", + "log.setLevel(logging.INFO)\n", + "logging.basicConfig()\n", + "log.info('%s logger started.', __name__)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:23.519815Z", + "start_time": "2017-08-05T23:33:23.299472Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "import argparse\n", + "import os\n", + "import sys\n", + "import gym\n", + "from gym import wrappers\n", + "import random\n", + "import numpy as np\n", + "\n", + "import torch\n", + "import torch.optim as optim\n", + "import torch.multiprocessing as mp\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "from torch.autograd import Variable" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:23.545021Z", + "start_time": "2017-08-05T23:33:23.521540Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "import os\n", + "os.sys.path.append(os.path.abspath('.'))\n", + "%reload_ext autoreload\n", + "%autoreload 2" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:23.587840Z", + "start_time": "2017-08-05T23:33:23.546733Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "'outputs/agent_portfolio-ddpo/2017-07-21_seperate_weights.pickle'" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "class Params():\n", + " def __init__(self):\n", + " # env\n", + " self.window_length = 50\n", + " # Model\n", + " self.batch_size = 250\n", + " self.lr = 3e-4\n", + " self.gamma = 0.00\n", + " self.gae_param = 0.95\n", + " self.clip = 0.2 # epsilon from eq 7, default 0.2\n", + " self.ent_coeff = 0.\n", + " self.num_epoch = 10\n", + " self.num_steps = 2048\n", + " self.time_horizon = 2000000\n", + " self.max_episode_length = 10000\n", + " self.seed = 1\n", + "\n", + "params = Params()\n", + "\n", + "save_path= 'outputs/agent_portfolio-ddpo/{}_seperate_weights.pickle'.format('2017-07-21')\n", + "try:\n", + " os.makedirs(os.path.dirname(save_path))\n", + "except OSError:\n", + " pass\n", + "save_path" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Memory\n", + "refs\n", + "- https://github.com/openai/baselines/blob/master/baselines/deepq/replay_buffer.py\n", + "- https://github.com/jaara/AI-blog/blob/master/Seaquest-DDQN-PER.py" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:23.620484Z", + "start_time": "2017-08-05T23:33:23.589508Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "class ReplayMemory(object):\n", + " def __init__(self, capacity):\n", + " self.capacity = capacity\n", + " self.memory = []\n", + "\n", + " def push(self, events):\n", + " for event in zip(*events):\n", + " self.memory.append(event)\n", + " if len(self.memory)>self.capacity:\n", + " del self.memory[0]\n", + "\n", + " def clear(self):\n", + " self.memory = []\n", + "\n", + " def sample(self, batch_size):\n", + " samples = zip(*random.sample(self.memory, batch_size))\n", + " return map(lambda x: torch.cat(x, 0), samples)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-02T00:55:29.885772Z", + "start_time": "2017-08-02T08:55:29.883459+08:00" + } + }, + "source": [ + "# Enviroment" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:23.954089Z", + "start_time": "2017-08-05T23:33:23.696823Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "(3, 5, 50)" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from src.environments.portfolio import PortfolioEnv, sharpe, max_drawdown\n", + "\n", + "# we want to pemute the channels a little\n", + "\n", + "class PermutedPortfolioEnv(PortfolioEnv):\n", + " def reset(self, *args, **kwargs):\n", + " return np.transpose(super().reset(*args, **kwargs),(0,1,2))\n", + " def step(self, *args, **kwargs):\n", + " observation, reward, done, info = super().step(*args, **kwargs)\n", + " observation = np.transpose(observation,(2,0,1))\n", + " return observation, reward, done, info\n", + "\n", + "\n", + "df_train = pd.read_hdf('./data/poloniex_30m.hf',key='train')\n", + "env = PermutedPortfolioEnv(\n", + " df=df_train,\n", + " steps=128, \n", + " scale=True, \n", + " augment=0.00025, # let just overfit first,\n", + " trading_cost=0, #0.0025, # let just overfit first,\n", + " window_length = params.window_length, \n", + ")\n", + "env.seed(params.seed)\n", + "env.reset().shape\n", + "\n", + "df_test = pd.read_hdf('./data/poloniex_30m.hf',key='test')\n", + "env_test = PermutedPortfolioEnv(\n", + " df=df_test,\n", + " steps=1280, \n", + " scale=True, \n", + " augment=0.00025, # let just overfit first,\n", + " trading_cost=0, #0.0025, # let just overfit first,\n", + " window_length = params.window_length, \n", + ")\n", + "env_test.seed(params.seed)\n", + "env_test.reset().shape" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-02T01:11:38.199434Z", + "start_time": "2017-08-02T09:11:38.155811+08:00" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:24.101770Z", + "start_time": "2017-08-05T23:33:23.955728Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "class GenericSharedModel(nn.Module):\n", + " def __init__(self, inputs, outputs):\n", + " super(GenericSharedModel, self).__init__()\n", + " num_inputs = int(np.prod(env.observation_space.shape))\n", + " num_outputs = int(np.prod(env.action_space.shape))\n", + " \n", + " # hidden layer sizes\n", + " h_size_1 = 64\n", + " h_size_2 = 64\n", + " \n", + " self.conv1 = nn.Conv2d(3, 2, (1, 3))\n", + " self.conv2 = nn.Conv2d(2, 20, (1, inputs[1] - 2))\n", + " \n", + " self.fc1 = nn.Linear(100, h_size_1)\n", + " self.fc2 = nn.Linear(h_size_1, h_size_2)\n", + " \n", + " self.mu = nn.Linear(h_size_2, num_outputs)\n", + " self.log_std = nn.Parameter(torch.zeros(num_outputs))\n", + " \n", + " \n", + " self.fc1b = nn.Linear(100, h_size_1)\n", + " self.fc2b = nn.Linear(h_size_1, h_size_2)\n", + " \n", + " self.v = nn.Linear(h_size_2,1)\n", + " \n", + " for name, p in self.named_parameters():\n", + " # init parameters\n", + " if 'bias' in name:\n", + " p.data.fill_(0)\n", + " '''\n", + " if 'mu.weight' in name:\n", + " p.data.normal_()\n", + " p.data /= torch.sum(p.data**2,0).expand_as(p.data)'''\n", + " \n", + " # mode\n", + " self.train()\n", + "\n", + " def forward(self, inputs):\n", + " # flatten\n", + "# inputs = inputs.view((inputs.size()[0],-1))\n", + " x = F.relu(self.conv1(inputs))\n", + " x = F.relu(self.conv2(x))\n", + " h = x.view(x.size(0),-1) # Flatten\n", + " \n", + " # actor\n", + " x = F.tanh(self.fc1(h))\n", + " x = F.tanh(self.fc2(x))\n", + " \n", + " # the action\n", + " mu = F.softmax(self.mu(x))\n", + " \n", + " # exploration multiplier\n", + " log_std = F.sigmoid(torch.exp(self.log_std).unsqueeze(0).expand_as(mu))\n", + " \n", + " # critic\n", + " x = F.tanh(self.fc1(h))\n", + " x = F.tanh(self.fc2(x))\n", + " v = self.v(x)\n", + " return mu, log_std, v" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Train" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:24.130649Z", + "start_time": "2017-08-05T23:33:24.103348Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "def mkdir(base, name):\n", + " path = os.path.join(base, name)\n", + " if not os.path.exists(path):\n", + " os.makedirs(path)\n", + " return path" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:24.179604Z", + "start_time": "2017-08-05T23:33:24.132353Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "\n", + "# class Shared_grad_buffers():\n", + "# def __init__(self, model):\n", + "# self.grads = {}\n", + "# for name, p in model.named_parameters():\n", + "# self.grads[name+'_grad'] = torch.ones(p.size()).share_memory_()\n", + "\n", + "# def add_gradient(self, model):\n", + "# for name, p in model.named_parameters():\n", + "# self.grads[name+'_grad'] += p.grad.data\n", + "\n", + "# def reset(self):\n", + "# for name,grad in self.grads.items():\n", + "# self.grads[name].fill_(0)\n", + "\n", + "class Shared_obs_stats():\n", + " \"\"\"Like batchnorm for input data\"\"\"\n", + " def __init__(self, num_inputs):\n", + " self.n = torch.zeros(num_inputs).share_memory_()\n", + " self.mean = torch.zeros(num_inputs).share_memory_()\n", + " self.mean_diff = torch.zeros(num_inputs).share_memory_()\n", + " self.var = torch.zeros(num_inputs).share_memory_()\n", + "\n", + " def observes(self, obs):\n", + " # observation mean var updates\n", + " x = obs.data.squeeze()\n", + " self.n += 1.\n", + " last_mean = self.mean.clone()\n", + " self.mean += (x-self.mean)/self.n\n", + " self.mean_diff += (x-last_mean)*(x-self.mean)\n", + " self.var = torch.clamp(self.mean_diff/self.n, min=1e-2)\n", + "\n", + " def normalize(self, inputs):\n", + " obs_mean = Variable(self.mean.unsqueeze(0).expand_as(inputs))\n", + " obs_std = Variable(torch.sqrt(self.var).unsqueeze(0).expand_as(inputs))\n", + " return torch.clamp((inputs-obs_mean)/obs_std, -5., 5.)" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:24.211410Z", + "start_time": "2017-08-05T23:33:24.181218Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "def normal(x, mu, sigma_sq):\n", + " a = (-1*(x-mu).pow(2)/(2*sigma_sq)).exp()\n", + " b = 1/(2*sigma_sq*np.pi).sqrt()\n", + " return a*b" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T23:33:24.248745Z", + "start_time": "2017-08-05T23:33:24.213002Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "GenericSharedModel (\n", + " (conv1): Conv2d(3, 2, kernel_size=(1, 3), stride=(1, 1))\n", + " (conv2): Conv2d(2, 20, kernel_size=(1, 48), stride=(1, 1))\n", + " (fc1): Linear (100 -> 64)\n", + " (fc2): Linear (64 -> 64)\n", + " (mu): Linear (64 -> 6)\n", + " (fc1b): Linear (100 -> 64)\n", + " (fc2b): Linear (64 -> 64)\n", + " (v): Linear (64 -> 1)\n", + ")" + ] + }, + "execution_count": 13, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "\n", + "cuda = False\n", + "torch.manual_seed(params.seed)\n", + "work_dir = mkdir('exp', 'ppo')\n", + "monitor_dir = mkdir(work_dir, 'monitor')\n", + "\n", + "# env = gym.make(params.env_name)\n", + "#env = wrappers.Monitor(env, monitor_dir, force=True)\n", + "\n", + "num_inputs = env.observation_space.shape[0]\n", + "num_outputs = env.action_space.shape[0]\n", + "\n", + "\n", + "#initialize network and optimizer\n", + "Model = GenericSharedModel\n", + "model = Model(env.observation_space.shape, env.action_space.shape)\n", + "if cuda: model.cuda()\n", + "\n", + "# shared_obs_stats = Shared_obs_stats(num_inputs)\n", + "optimizer = optim.Adam(model.parameters(), lr=params.lr)\n", + "model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-04T02:50:27.128726Z", + "start_time": "2017-08-04T10:50:01.031402+08:00" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T04:13:28.173675Z", + "start_time": "2017-08-05T23:33:24.250420Z" + }, + "scrolled": true + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Widget Javascript not detected. It may not be installed or enabled properly.\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "f2fee12f88094233aa0076e1ac1bb4a7" + } + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=3.477e-05, reward=3.824e-08, cash_bias=0.2268, portfolio_value=1.021, market_value=1.035, episode=15\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/wassname/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/torch/serialization.py:147: UserWarning: Couldn't retrieve source code for container of type GenericSharedModel. It won't be checked for correctness upon loading.\n", + " \"type \" + obj.__name__ + \". It won't be checked \"\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=0.0001077, reward=-6.766e-06, cash_bias=0.0897, portfolio_value=1.012, market_value=1.006, episode=31\n", + "loss=1.51e-05, reward=1.044e-05, cash_bias=0.1304, portfolio_value=1.058, market_value=1.038, episode=47\n", + "loss=1.373e-06, reward=-1.433e-05, cash_bias=0.156, portfolio_value=1.02, market_value=1.034, episode=63\n", + "loss=3.54e-07, reward=3.544e-06, cash_bias=0.1662, portfolio_value=1.048, market_value=1.04, episode=79\n", + "loss=1.341e-07, reward=-4.519e-06, cash_bias=0.1244, portfolio_value=1.037, market_value=1.037, episode=95\n", + "loss=4.816e-08, reward=-6.312e-06, cash_bias=0.1355, portfolio_value=1.03, market_value=1.022, episode=111\n", + "loss=1.196e-07, reward=3.038e-05, cash_bias=0.1946, portfolio_value=1.093, market_value=1.101, episode=127\n", + "loss=3.851e-07, reward=-9.574e-06, cash_bias=0.1422, portfolio_value=1.033, market_value=1.029, episode=143\n", + "loss=3.019e-07, reward=-1.495e-05, cash_bias=0.174, portfolio_value=1.049, market_value=1.063, episode=159\n", + "loss=5.147e-07, reward=1.896e-05, cash_bias=0.09807, portfolio_value=1.055, market_value=1.049, episode=175\n", + "loss=9.991e-07, reward=1.223e-05, cash_bias=0.181, portfolio_value=1.034, market_value=1.045, episode=191\n", + "loss=2.164e-06, reward=1.334e-05, cash_bias=0.126, portfolio_value=1.058, market_value=1.051, episode=207\n", + "loss=1.113e-06, reward=5.948e-06, cash_bias=0.1825, portfolio_value=1.017, market_value=1.048, episode=223\n", + "loss=1.739e-07, reward=1.045e-05, cash_bias=0.1644, portfolio_value=1.033, market_value=1.047, episode=239\n", + "loss=4.251e-07, reward=1.203e-05, cash_bias=0.1335, portfolio_value=1.058, market_value=1.046, episode=255\n", + "loss=2.202e-07, reward=5.999e-06, cash_bias=0.1481, portfolio_value=1.041, market_value=1.029, episode=271\n", + "loss=7.05e-08, reward=-2.534e-05, cash_bias=0.1113, portfolio_value=1.067, market_value=1.043, episode=287\n", + "loss=3.176e-08, reward=1.078e-05, cash_bias=0.1876, portfolio_value=1.017, market_value=1.042, episode=303\n", + "loss=5.738e-08, reward=-6.769e-06, cash_bias=0.1628, portfolio_value=1.056, market_value=1.041, episode=319\n", + "loss=6.528e-08, reward=-2.196e-06, cash_bias=0.1909, portfolio_value=1.016, market_value=1.039, episode=335\n", + "loss=6.426e-08, reward=-7.104e-09, cash_bias=0.1092, portfolio_value=1.031, market_value=1.026, episode=351\n", + "loss=1.037e-07, reward=-4.56e-06, cash_bias=0.1919, portfolio_value=1.043, market_value=1.026, episode=367\n", + "loss=9.729e-08, reward=1.071e-05, cash_bias=0.1233, portfolio_value=1.086, market_value=1.067, episode=383\n", + "loss=4.405e-07, reward=7.087e-06, cash_bias=0.1403, portfolio_value=1.006, market_value=1.002, episode=399\n", + "loss=2.907e-07, reward=1.393e-05, cash_bias=0.1337, portfolio_value= 1, market_value=1.005, episode=415\n", + "loss=5.824e-08, reward=-2.914e-05, cash_bias=0.2066, portfolio_value=1.026, market_value=1.036, episode=431\n", + "loss=6.702e-08, reward=-3.884e-06, cash_bias=0.208, portfolio_value=1.058, market_value=1.029, episode=447\n", + "loss=7.249e-08, reward=-1.914e-05, cash_bias=0.2516, portfolio_value=1.03, market_value=1.049, episode=463\n", + "loss=4.019e-08, reward=2.214e-05, cash_bias=0.09475, portfolio_value=1.032, market_value=1.042, episode=479\n", + "loss=4.103e-08, reward=4.737e-06, cash_bias=0.07838, portfolio_value=1.077, market_value=1.053, episode=495\n", + "loss=6.7e-08, reward=5.126e-06, cash_bias=0.2116, portfolio_value=1.023, market_value=1.021, episode=511\n", + "loss=3.269e-07, reward=-4.624e-06, cash_bias=0.2729, portfolio_value=1.028, market_value=1.03, episode=527\n", + "loss=7.444e-07, reward=-1.383e-05, cash_bias=0.1911, portfolio_value=1.008, market_value=1.014, episode=543\n", + "loss=1.982e-06, reward=-1.426e-05, cash_bias=0.1706, portfolio_value=1.059, market_value=1.026, episode=559\n", + "loss=6.095e-06, reward=1.607e-05, cash_bias=0.1887, portfolio_value=1.027, market_value=1.026, episode=575\n", + "loss=5.246e-06, reward=4.424e-06, cash_bias=0.1476, portfolio_value=0.9932, market_value=0.9963, episode=591\n", + "loss=8.779e-06, reward=2e-05, cash_bias=0.1574, portfolio_value=1.032, market_value=1.036, episode=607\n", + "loss=1.717e-05, reward=3.525e-06, cash_bias=0.1789, portfolio_value=1.022, market_value=1.016, episode=623\n", + "loss=4.484e-05, reward=3.999e-05, cash_bias=0.1907, portfolio_value=1.001, market_value=1.006, episode=639\n", + "loss=0.0001553, reward=8.762e-06, cash_bias=0.2686, portfolio_value=1.019, market_value=1.019, episode=655\n", + "loss=6.204e-06, reward=1.851e-05, cash_bias=0.174, portfolio_value=1.057, market_value=1.03, episode=671\n", + "loss=2.001e-05, reward=1.829e-05, cash_bias=0.2124, portfolio_value=1.05, market_value=1.042, episode=687\n", + "loss=5.237e-06, reward=-1.892e-05, cash_bias=0.216, portfolio_value=1.031, market_value=1.024, episode=703\n", + "loss=6.333e-06, reward=7.618e-06, cash_bias=0.1945, portfolio_value=1.036, market_value=1.021, episode=719\n", + "loss=6.736e-07, reward=3.046e-05, cash_bias=0.2737, portfolio_value=1.054, market_value=1.05, episode=735\n", + "loss=1.632e-06, reward=1.172e-05, cash_bias=0.1065, portfolio_value=1.045, market_value=1.042, episode=751\n", + "loss=4.664e-07, reward=-1.698e-05, cash_bias=0.1918, portfolio_value=1.026, market_value=1.033, episode=767\n", + "loss=3.534e-08, reward=1.895e-06, cash_bias=0.136, portfolio_value=1.03, market_value=1.042, episode=783\n", + "loss=1.149e-07, reward=1.773e-05, cash_bias=0.1871, portfolio_value=1.046, market_value=1.053, episode=799\n", + "loss=9.931e-08, reward=7.151e-06, cash_bias=0.1598, portfolio_value=1.006, market_value=1.017, episode=815\n", + "loss=1.042e-07, reward=-7.497e-06, cash_bias=0.007246, portfolio_value=1.071, market_value=1.048, episode=831\n", + "loss=1.583e-07, reward=-2.322e-05, cash_bias=0.1571, portfolio_value=1.028, market_value=1.028, episode=847\n", + "loss=1.588e-07, reward=-1.984e-05, cash_bias=0.07928, portfolio_value=1.081, market_value=1.067, episode=863\n", + "loss=1.714e-07, reward=-1.2e-05, cash_bias=0.1303, portfolio_value=1.053, market_value=1.018, episode=879\n", + "loss=1.723e-07, reward=-6.748e-06, cash_bias=0.1795, portfolio_value=1.027, market_value=1.033, episode=895\n", + "loss=1.51e-07, reward=7.896e-06, cash_bias=0.2554, portfolio_value=1.007, market_value=1.009, episode=911\n", + "loss=1.543e-07, reward=-8.116e-06, cash_bias=0.2124, portfolio_value=1.014, market_value=1.039, episode=927\n", + "loss=1.772e-07, reward=-1.878e-05, cash_bias=0.2859, portfolio_value=1.029, market_value=1.027, episode=943\n", + "loss=3.28e-07, reward=-1.063e-05, cash_bias=0.2929, portfolio_value=1.023, market_value=1.029, episode=959\n", + "loss=1.4e-07, reward=-4.049e-06, cash_bias=0.1373, portfolio_value=1.068, market_value=1.079, episode=975\n", + "loss=7.534e-08, reward=-1.822e-05, cash_bias=0.2066, portfolio_value=0.994, market_value=1.008, episode=991\n", + "loss=3.886e-08, reward=-8.5e-06, cash_bias=0.1413, portfolio_value=1.057, market_value=1.059, episode=1007\n", + "loss=3.407e-08, reward=3.611e-05, cash_bias=0.2074, portfolio_value=1.046, market_value=1.04, episode=1023\n", + "loss=1.826e-08, reward=4.325e-06, cash_bias=0.1639, portfolio_value=1.081, market_value=1.078, episode=1039\n", + "loss=1.751e-08, reward=-3.702e-06, cash_bias=0.143, portfolio_value=1.032, market_value=1.03, episode=1055\n", + "loss=1.511e-08, reward=2.034e-06, cash_bias=0.2087, portfolio_value=1.053, market_value=1.062, episode=1071\n", + "loss=3e-08, reward=6.775e-06, cash_bias=0.2016, portfolio_value=1.047, market_value=1.018, episode=1087\n", + "loss=1.918e-08, reward=8.09e-06, cash_bias=0.2075, portfolio_value= 1, market_value=1.024, episode=1103\n", + "loss=2.357e-08, reward=2.91e-05, cash_bias=0.2053, portfolio_value=0.9971, market_value=1.031, episode=1119\n", + "loss=2.511e-08, reward=1.029e-06, cash_bias=0.2139, portfolio_value=1.022, market_value=1.018, episode=1135\n", + "loss=1.602e-08, reward=-1.196e-05, cash_bias=0.1998, portfolio_value=0.9944, market_value=1.002, episode=1151\n", + "loss=2.526e-08, reward=-2.624e-06, cash_bias=0.1345, portfolio_value=1.057, market_value=1.04, episode=1167\n", + "loss=2.028e-08, reward=-9.411e-06, cash_bias=0.2107, portfolio_value=1.026, market_value=1.048, episode=1183\n", + "loss=1.371e-08, reward=-2.725e-05, cash_bias=0.09784, portfolio_value=1.054, market_value=1.062, episode=1199\n", + "loss=1.506e-08, reward=1.964e-06, cash_bias=0.0989, portfolio_value=1.077, market_value=1.071, episode=1215\n", + "loss=3.64e-08, reward=-2.613e-05, cash_bias=0.2084, portfolio_value=1.022, market_value=1.031, episode=1231\n", + "loss=2.582e-08, reward=1.924e-05, cash_bias=0.1523, portfolio_value=1.055, market_value=1.062, episode=1247\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=1.278e-08, reward=1.017e-05, cash_bias=0.06298, portfolio_value=1.071, market_value=1.047, episode=1263\n", + "loss=2.192e-08, reward=1.447e-05, cash_bias=0.06402, portfolio_value=1.042, market_value=1.034, episode=1279\n", + "loss=2.989e-08, reward=1.194e-05, cash_bias=0.1197, portfolio_value=1.059, market_value=1.048, episode=1295\n", + "loss=1.702e-08, reward=4.84e-06, cash_bias=0.07974, portfolio_value=1.044, market_value=1.052, episode=1311\n", + "loss=1.918e-08, reward=-1.603e-05, cash_bias=0.1815, portfolio_value=1.036, market_value=1.041, episode=1327\n", + "loss=2.345e-08, reward=6.147e-06, cash_bias=0.249, portfolio_value=1.006, market_value=1.02, episode=1343\n", + "loss=3.845e-08, reward=-1.46e-05, cash_bias=0.1187, portfolio_value=0.9999, market_value=1.008, episode=1359\n", + "loss=1.17e-08, reward=2.131e-05, cash_bias=0.1551, portfolio_value=1.036, market_value=1.024, episode=1375\n", + "loss=2.48e-08, reward=7.592e-06, cash_bias=0.1231, portfolio_value=1.079, market_value=1.057, episode=1391\n", + "loss=1.491e-07, reward=-7.406e-05, cash_bias=0.155, portfolio_value=1.035, market_value=1.015, episode=1407\n", + "loss=3.674e-08, reward=-7.698e-06, cash_bias=0.1663, portfolio_value=1.035, market_value=1.019, episode=1423\n", + "loss=1.689e-08, reward=-3.651e-05, cash_bias=0.2155, portfolio_value=1.032, market_value=1.003, episode=1439\n", + "loss=1.203e-08, reward=5.654e-06, cash_bias=0.1418, portfolio_value=1.05, market_value=1.043, episode=1455\n", + "loss=1.695e-08, reward=-9.818e-06, cash_bias=0.118, portfolio_value=1.001, market_value=1.048, episode=1471\n", + "loss=3.06e-08, reward=3.18e-06, cash_bias=0.1032, portfolio_value=0.9808, market_value=1.023, episode=1487\n", + "loss=3.897e-08, reward=-7.85e-06, cash_bias=0.2443, portfolio_value=1.035, market_value=1.045, episode=1503\n", + "loss=4.199e-08, reward=1.06e-05, cash_bias=0.1825, portfolio_value=1.065, market_value=1.035, episode=1519\n", + "loss=5.68e-08, reward=-3.342e-06, cash_bias=0.194, portfolio_value=1.039, market_value=1.023, episode=1535\n", + "loss=9.116e-08, reward=9.549e-06, cash_bias=0.1197, portfolio_value=1.046, market_value=1.034, episode=1551\n", + "loss=1.834e-07, reward=4.303e-06, cash_bias=0.2255, portfolio_value=1.053, market_value=1.045, episode=1567\n", + "loss=1.803e-07, reward=9.106e-06, cash_bias=0.2581, portfolio_value=1.023, market_value=1.034, episode=1583\n", + "loss=2.363e-07, reward=-2.656e-05, cash_bias=0.1758, portfolio_value=1.026, market_value=1.04, episode=1599\n", + "loss=2.152e-07, reward=-1.212e-05, cash_bias=0.166, portfolio_value=1.034, market_value=1.02, episode=1615\n", + "loss=3.246e-07, reward=-2.673e-06, cash_bias=0.1357, portfolio_value=1.03, market_value=1.06, episode=1631\n", + "loss=1.19e-06, reward=-2.366e-05, cash_bias=0.219, portfolio_value=1.011, market_value=1.023, episode=1647\n", + "loss=4.484e-06, reward=7.195e-06, cash_bias=0.1359, portfolio_value=1.027, market_value=1.018, episode=1663\n", + "loss=1.077e-05, reward=1.109e-05, cash_bias=0.163, portfolio_value=1.004, market_value=1.036, episode=1679\n", + "loss=7.597e-06, reward=-8.753e-06, cash_bias=0.1795, portfolio_value=1.095, market_value=1.06, episode=1695\n", + "loss=1.289e-05, reward=1.854e-05, cash_bias=0.1318, portfolio_value=1.03, market_value=1.04, episode=1711\n", + "loss=1.99e-05, reward=8.154e-06, cash_bias=0.1168, portfolio_value=1.021, market_value=1.04, episode=1727\n", + "loss=1.959e-05, reward=3.023e-05, cash_bias=0.06793, portfolio_value=1.023, market_value=1.034, episode=1743\n", + "loss=2.125e-06, reward=-4.968e-06, cash_bias=0.1762, portfolio_value=1.064, market_value=1.053, episode=1759\n", + "loss=2.243e-06, reward=1.765e-05, cash_bias=0.1363, portfolio_value=1.036, market_value=1.036, episode=1775\n", + "loss=1.951e-06, reward=1.991e-06, cash_bias=0.2337, portfolio_value=1.033, market_value=1.046, episode=1791\n", + "loss=4.308e-07, reward=-8.736e-06, cash_bias=0.1507, portfolio_value=1.028, market_value=1.043, episode=1807\n", + "loss=4.976e-07, reward=1.902e-05, cash_bias=0.1705, portfolio_value=1.08, market_value=1.06, episode=1823\n", + "loss=5.078e-07, reward=-7.406e-07, cash_bias=0.2543, portfolio_value=1.024, market_value=1.038, episode=1839\n", + "loss=7.619e-07, reward=6.263e-06, cash_bias=0.1606, portfolio_value=1.037, market_value=1.028, episode=1855\n", + "loss=1.941e-06, reward=3.325e-06, cash_bias=0.101, portfolio_value=1.05, market_value=1.029, episode=1871\n", + "loss=1.508e-06, reward=4.439e-06, cash_bias=0.2207, portfolio_value=0.9936, market_value=1.007, episode=1887\n", + "loss=1.288e-07, reward=-3.108e-05, cash_bias=0.1768, portfolio_value=1.054, market_value=1.054, episode=1903\n", + "loss=8.136e-07, reward=-1.089e-05, cash_bias=0.2333, portfolio_value=1.051, market_value=1.044, episode=1919\n", + "loss=2.544e-06, reward=1.693e-05, cash_bias=0.1163, portfolio_value=1.031, market_value=1.042, episode=1935\n", + "loss=2.353e-05, reward=3.191e-06, cash_bias=0.1346, portfolio_value=1.039, market_value=1.036, episode=1951\n", + "loss=0.0001192, reward=1.238e-05, cash_bias=0.1557, portfolio_value=1.072, market_value=1.053, episode=1967\n", + "loss=2.15e-05, reward=9.444e-06, cash_bias=0.1141, portfolio_value=1.055, market_value=1.057, episode=1983\n", + "loss=0.0001428, reward=3.915e-06, cash_bias=0.1487, portfolio_value=1.053, market_value=1.036, episode=1999\n", + "loss=0.0003267, reward=-3.643e-06, cash_bias=0.1345, portfolio_value=1.016, market_value=1.043, episode=2015\n", + "loss=0.000828, reward=1.878e-06, cash_bias=0.2057, portfolio_value=1.038, market_value=1.03, episode=2031\n", + "loss=0.0001765, reward=8.763e-06, cash_bias=0.1135, portfolio_value=1.027, market_value=1.036, episode=2047\n", + "loss=5.094e-05, reward=-1.717e-06, cash_bias=0.08979, portfolio_value=1.048, market_value=1.038, episode=2063\n", + "loss=2.066e-05, reward=1.875e-05, cash_bias=0.1773, portfolio_value=1.019, market_value=1.017, episode=2079\n", + "loss=2.23e-05, reward=-9.974e-06, cash_bias=0.1755, portfolio_value=1.039, market_value=1.03, episode=2095\n", + "loss=2.788e-05, reward=7.176e-06, cash_bias=0.2036, portfolio_value=1.078, market_value=1.053, episode=2111\n", + "loss=2.015e-05, reward=1.265e-05, cash_bias=0.1403, portfolio_value=1.093, market_value=1.05, episode=2127\n", + "loss=1.83e-05, reward=-1.916e-05, cash_bias=0.07565, portfolio_value=1.073, market_value=1.059, episode=2143\n", + "loss=2.663e-05, reward=5.219e-06, cash_bias=0.09925, portfolio_value=1.03, market_value=1.013, episode=2159\n", + "loss=3.777e-05, reward=4.894e-05, cash_bias=0.1157, portfolio_value=1.062, market_value=1.035, episode=2175\n", + "loss=3.634e-05, reward=4.816e-06, cash_bias=0.1484, portfolio_value=1.051, market_value=1.042, episode=2191\n", + "loss=2.539e-05, reward=4.987e-06, cash_bias=0.158, portfolio_value=1.014, market_value=1.01, episode=2207\n", + "loss=3.184e-05, reward=-7.66e-07, cash_bias=0.2408, portfolio_value=1.024, market_value=1.035, episode=2223\n", + "loss=4.311e-05, reward=3.19e-06, cash_bias=0.1716, portfolio_value=1.043, market_value=1.029, episode=2239\n", + "loss=5.731e-05, reward=-2.268e-06, cash_bias=0.154, portfolio_value=1.013, market_value=1.022, episode=2255\n", + "loss=6.752e-05, reward=3.766e-06, cash_bias=0.2279, portfolio_value=1.016, market_value=1.023, episode=2271\n", + "loss=7.741e-05, reward=-7.202e-06, cash_bias=0.1772, portfolio_value=1.024, market_value=1.01, episode=2287\n", + "loss=8.258e-05, reward=-1.083e-05, cash_bias=0.1825, portfolio_value=1.019, market_value=1.03, episode=2303\n", + "loss=5.76e-05, reward=-5.237e-06, cash_bias=0.1816, portfolio_value=1.058, market_value=1.062, episode=2319\n", + "loss=3.129e-05, reward=4.805e-06, cash_bias=0.1931, portfolio_value=1.028, market_value=1.048, episode=2335\n", + "loss=2.661e-05, reward=-3.591e-06, cash_bias=0.1008, portfolio_value=1.051, market_value=1.033, episode=2351\n", + "loss=2.872e-05, reward=2.905e-06, cash_bias=0.2087, portfolio_value=1.041, market_value=1.069, episode=2367\n", + "loss=2.984e-05, reward=6.993e-06, cash_bias=0.2122, portfolio_value=1.044, market_value=1.045, episode=2383\n", + "loss=2.345e-05, reward=6.959e-06, cash_bias=0.1568, portfolio_value=1.081, market_value=1.07, episode=2399\n", + "loss=2.164e-05, reward=3.03e-05, cash_bias=0.175, portfolio_value=1.051, market_value=1.055, episode=2415\n", + "loss=2.391e-05, reward=-6.063e-06, cash_bias=0.1339, portfolio_value=1.088, market_value=1.067, episode=2431\n", + "loss=2.318e-05, reward=-2.068e-05, cash_bias=0.1047, portfolio_value=0.989, market_value=1.024, episode=2447\n", + "loss=2.422e-05, reward=-1.993e-06, cash_bias=0.2375, portfolio_value=1.051, market_value=1.054, episode=2463\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=3.084e-05, reward=5.34e-06, cash_bias=0.124, portfolio_value=1.052, market_value=1.038, episode=2479\n", + "loss=3.316e-05, reward=1.018e-06, cash_bias=0.1926, portfolio_value=1.013, market_value=1.031, episode=2495\n", + "loss=2.601e-05, reward=5.617e-06, cash_bias=0.1381, portfolio_value=0.9987, market_value=1.03, episode=2511\n", + "loss=1.975e-05, reward=3.413e-05, cash_bias=0.07923, portfolio_value=1.062, market_value=1.051, episode=2527\n", + "loss=1.696e-05, reward=-5.929e-07, cash_bias=0.2143, portfolio_value=1.006, market_value=1.013, episode=2543\n", + "loss=2.249e-05, reward=-6.087e-06, cash_bias=0.1697, portfolio_value=1.014, market_value=1.023, episode=2559\n", + "loss=2.485e-05, reward=4.919e-06, cash_bias=0.1799, portfolio_value=1.034, market_value=1.032, episode=2575\n", + "loss=1.959e-05, reward=-1.292e-06, cash_bias=0.1624, portfolio_value=1.036, market_value=1.017, episode=2591\n", + "loss=1.548e-05, reward=-9.585e-06, cash_bias=0.1253, portfolio_value=1.034, market_value=1.047, episode=2607\n", + "loss=1.417e-05, reward=4.047e-06, cash_bias=0.1815, portfolio_value=1.035, market_value=1.054, episode=2623\n", + "loss=1.562e-05, reward=-4.324e-06, cash_bias=0.1293, portfolio_value=1.062, market_value=1.04, episode=2639\n", + "loss=1.81e-05, reward=-3.081e-06, cash_bias=0.1504, portfolio_value=1.051, market_value=1.046, episode=2655\n", + "loss=2.492e-05, reward=1.686e-05, cash_bias=0.1329, portfolio_value=1.051, market_value=1.029, episode=2671\n", + "loss=2.79e-05, reward=-3.096e-05, cash_bias=0.2663, portfolio_value=1.031, market_value=1.025, episode=2687\n", + "loss=3.302e-05, reward=1.976e-05, cash_bias=0.14, portfolio_value=1.029, market_value=1.047, episode=2703\n", + "loss=3.765e-05, reward=-1.338e-05, cash_bias=0.1512, portfolio_value=1.019, market_value=1.034, episode=2719\n", + "loss=4.78e-05, reward=-7.534e-07, cash_bias=0.2561, portfolio_value=1.017, market_value=1.024, episode=2735\n", + "loss=8.136e-05, reward=-6.883e-06, cash_bias=0.1216, portfolio_value=1.027, market_value=1.034, episode=2751\n", + "loss=7.077e-05, reward=-3.751e-06, cash_bias=0.1605, portfolio_value=1.025, market_value=1.021, episode=2767\n", + "loss=4.393e-05, reward=1.681e-05, cash_bias=0.2061, portfolio_value=1.019, market_value=1.009, episode=2783\n", + "loss=2.534e-05, reward=-1.397e-05, cash_bias=0.1271, portfolio_value=1.03, market_value=1.03, episode=2799\n", + "loss=1.962e-05, reward=3.258e-06, cash_bias=0.1699, portfolio_value=1.031, market_value=1.043, episode=2815\n", + "loss=1.591e-05, reward=2.694e-05, cash_bias=0.1809, portfolio_value=1.036, market_value=1.028, episode=2831\n", + "loss=9.061e-06, reward=-1.147e-07, cash_bias=0.1657, portfolio_value=1.023, market_value=1.026, episode=2847\n", + "loss=8.621e-06, reward=2.264e-07, cash_bias=0.31, portfolio_value=1.055, market_value=1.049, episode=2863\n", + "loss=9.411e-06, reward=6.386e-07, cash_bias=0.1509, portfolio_value=1.003, market_value=1.024, episode=2879\n", + "loss=1.032e-05, reward=-9.789e-06, cash_bias=0.2008, portfolio_value=0.9977, market_value=1.008, episode=2895\n", + "loss=1.172e-05, reward=6.231e-06, cash_bias=0.1283, portfolio_value=0.997, market_value=1.046, episode=2911\n", + "loss=1.37e-05, reward=-2.903e-06, cash_bias=0.1039, portfolio_value=1.048, market_value=1.028, episode=2927\n", + "loss=1.865e-05, reward=1.601e-05, cash_bias=0.1326, portfolio_value=0.9876, market_value=1.006, episode=2943\n", + "loss=2.074e-05, reward=8.569e-06, cash_bias=0.225, portfolio_value=0.9758, market_value=1.008, episode=2959\n", + "loss=2.256e-05, reward=-2.326e-06, cash_bias=0.2863, portfolio_value=1.041, market_value=1.03, episode=2975\n", + "loss=1.402e-05, reward=-5.343e-06, cash_bias=0.1034, portfolio_value=1.047, market_value=1.036, episode=2991\n", + "loss=3.046e-06, reward=7.114e-07, cash_bias=0.1883, portfolio_value=1.089, market_value=1.075, episode=3007\n", + "loss=1.161e-07, reward=-1.061e-05, cash_bias=0.1243, portfolio_value=1.013, market_value=1.019, episode=3023\n", + "loss=2.23e-07, reward=-1.172e-06, cash_bias=0.1814, portfolio_value=1.026, market_value=1.007, episode=3039\n", + "loss=4.325e-07, reward=-5.738e-06, cash_bias=0.14, portfolio_value=1.047, market_value=1.041, episode=3055\n", + "loss=5.119e-07, reward=1.367e-05, cash_bias=0.06532, portfolio_value=1.073, market_value=1.065, episode=3071\n", + "loss=5.888e-07, reward=2.595e-05, cash_bias=0.2157, portfolio_value=1.033, market_value=1.012, episode=3087\n", + "loss=5.1e-07, reward=1.191e-05, cash_bias=0.174, portfolio_value=0.9998, market_value=1.042, episode=3103\n", + "loss=4.985e-07, reward=2.406e-05, cash_bias=0.2227, portfolio_value=1.056, market_value=1.044, episode=3119\n", + "loss=6.435e-07, reward=-9.77e-07, cash_bias=0.1259, portfolio_value=1.008, market_value=1.036, episode=3135\n", + "loss=8.363e-07, reward=-6.327e-06, cash_bias=0.228, portfolio_value=1.025, market_value=1.024, episode=3151\n", + "loss=1.554e-06, reward=5.536e-06, cash_bias=0.1305, portfolio_value=1.018, market_value=0.9981, episode=3167\n", + "loss=1.036e-06, reward=-1.623e-05, cash_bias=0.1946, portfolio_value=1.006, market_value=1.044, episode=3183\n", + "loss=1.003e-06, reward=3.882e-06, cash_bias=0.1549, portfolio_value=1.036, market_value=1.025, episode=3199\n", + "loss=8.574e-07, reward=4.078e-06, cash_bias=0.2459, portfolio_value=1.044, market_value=1.061, episode=3215\n", + "loss=9.465e-07, reward=-1.713e-05, cash_bias=0.184, portfolio_value=1.039, market_value=1.036, episode=3231\n", + "loss=1.008e-06, reward=6.902e-06, cash_bias=0.2004, portfolio_value=0.9995, market_value=1.045, episode=3247\n", + "loss=9.924e-07, reward=1.714e-06, cash_bias=0.144, portfolio_value=1.039, market_value=1.019, episode=3263\n", + "loss=7.175e-07, reward=-1.807e-05, cash_bias=0.2078, portfolio_value=1.069, market_value=1.054, episode=3279\n", + "loss=5.355e-07, reward=-1.311e-06, cash_bias=0.1628, portfolio_value=1.027, market_value=1.014, episode=3295\n", + "loss=3.383e-07, reward=4.842e-06, cash_bias=0.1357, portfolio_value=1.064, market_value=1.044, episode=3311\n", + "loss=2.332e-07, reward=1.413e-05, cash_bias=0.1585, portfolio_value=1.032, market_value=1.046, episode=3327\n", + "loss=1.706e-07, reward=-1.139e-05, cash_bias=0.1621, portfolio_value=1.027, market_value=1.021, episode=3343\n", + "loss=1.159e-07, reward=5.069e-06, cash_bias=0.1394, portfolio_value=1.041, market_value=1.029, episode=3359\n", + "loss=8.385e-08, reward=2.277e-05, cash_bias=0.1246, portfolio_value=1.077, market_value=1.056, episode=3375\n", + "loss=6.655e-08, reward=5.81e-06, cash_bias=0.1245, portfolio_value=1.074, market_value=1.063, episode=3391\n", + "loss=7.904e-08, reward=-4.141e-06, cash_bias=0.1144, portfolio_value=1.03, market_value=1.016, episode=3407\n", + "loss=8.728e-08, reward=2.459e-05, cash_bias=0.2322, portfolio_value=1.017, market_value=1.03, episode=3423\n", + "loss=9.359e-08, reward=2.636e-06, cash_bias=0.1744, portfolio_value=1.045, market_value=1.063, episode=3439\n", + "loss=6.302e-08, reward=-6.022e-06, cash_bias=0.1513, portfolio_value=1.029, market_value=1.048, episode=3455\n", + "loss=4.052e-08, reward=1.019e-05, cash_bias=0.1766, portfolio_value=1.022, market_value=1.036, episode=3471\n", + "loss=5.366e-08, reward=3.769e-05, cash_bias=0.1131, portfolio_value=1.042, market_value=1.049, episode=3487\n", + "loss=4.929e-08, reward=-1.33e-05, cash_bias=0.1831, portfolio_value=1.028, market_value=1.048, episode=3503\n", + "loss=7.695e-08, reward=2.207e-06, cash_bias=0.1337, portfolio_value=1.064, market_value=1.054, episode=3519\n", + "loss=8.856e-08, reward=-5.282e-06, cash_bias=0.2536, portfolio_value=1.019, market_value=1.02, episode=3535\n", + "loss=7.316e-08, reward=-1.01e-05, cash_bias=0.1679, portfolio_value=1.033, market_value=1.057, episode=3551\n", + "loss=5.78e-08, reward=-2.599e-06, cash_bias=0.1086, portfolio_value=1.046, market_value=1.034, episode=3567\n", + "loss=4.42e-08, reward=-6.114e-06, cash_bias=0.1191, portfolio_value=1.001, market_value=1.017, episode=3583\n", + "loss=5.231e-08, reward=5.195e-06, cash_bias=0.249, portfolio_value=1.036, market_value=1.027, episode=3599\n", + "loss=3.996e-08, reward=-2.77e-05, cash_bias=0.1214, portfolio_value=1.044, market_value=1.035, episode=3615\n", + "loss=3.416e-08, reward=-1.134e-05, cash_bias=0.1504, portfolio_value=1.037, market_value=1.05, episode=3631\n", + "loss=3.771e-08, reward=6.355e-06, cash_bias=0.1421, portfolio_value=1.045, market_value=1.036, episode=3647\n", + "loss=4.708e-08, reward=1.067e-06, cash_bias=0.2273, portfolio_value=1.027, market_value=1.018, episode=3663\n", + "loss=1.207e-07, reward=-1.1e-05, cash_bias=0.2102, portfolio_value=1.085, market_value=1.079, episode=3679\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=4.387e-08, reward=-2.352e-05, cash_bias=0.1878, portfolio_value=1.037, market_value=1.018, episode=3695\n", + "loss=4.384e-08, reward=3.938e-06, cash_bias=0.14, portfolio_value=0.9945, market_value=1.024, episode=3711\n", + "loss=4.917e-08, reward=1.023e-05, cash_bias=0.1052, portfolio_value=1.011, market_value=1.028, episode=3727\n", + "loss=5.426e-08, reward=1.41e-05, cash_bias=0.1996, portfolio_value=1.026, market_value=1.04, episode=3743\n", + "loss=5.895e-08, reward=4.465e-06, cash_bias=0.1876, portfolio_value=1.08, market_value=1.067, episode=3759\n", + "loss=6.845e-08, reward=-1.859e-05, cash_bias=0.1701, portfolio_value=1.063, market_value=1.053, episode=3775\n", + "loss=6.106e-08, reward=-1.347e-05, cash_bias=0.1553, portfolio_value=1.038, market_value=1.038, episode=3791\n", + "loss=6.527e-08, reward=-2.678e-06, cash_bias=0.2475, portfolio_value=1.008, market_value=1.002, episode=3807\n", + "loss=8.147e-08, reward=-2.672e-06, cash_bias=0.09328, portfolio_value=1.02, market_value=1.044, episode=3823\n", + "loss=8.616e-08, reward=1.856e-05, cash_bias=0.1996, portfolio_value=1.035, market_value=1.024, episode=3839\n", + "loss=8.852e-08, reward=6.742e-06, cash_bias=0.1274, portfolio_value=1.027, market_value=1.022, episode=3855\n", + "loss=9.316e-08, reward=2.157e-05, cash_bias=0.2656, portfolio_value=1.011, market_value=1.014, episode=3871\n", + "loss=2.021e-07, reward=-2.412e-06, cash_bias=0.2012, portfolio_value=1.003, market_value=1.023, episode=3887\n", + "loss=1.212e-07, reward=1.908e-05, cash_bias=0.2352, portfolio_value=1.02, market_value=1.011, episode=3903\n", + "loss=1.172e-07, reward=2.26e-05, cash_bias=0.1751, portfolio_value=1.063, market_value=1.054, episode=3919\n", + "loss=1.122e-07, reward=1.01e-05, cash_bias=0.3032, portfolio_value=1.03, market_value=1.035, episode=3935\n", + "loss=9.183e-08, reward=-7.24e-06, cash_bias=0.1502, portfolio_value=1.015, market_value=1.041, episode=3951\n", + "loss=7.33e-08, reward=2.661e-06, cash_bias=0.1245, portfolio_value=1.056, market_value=1.062, episode=3967\n", + "loss=5.374e-08, reward=6.836e-06, cash_bias=0.182, portfolio_value=1.022, market_value=0.9997, episode=3983\n", + "loss=6.105e-08, reward=1.187e-05, cash_bias=0.1511, portfolio_value=1.014, market_value=0.9898, episode=3999\n", + "loss=4.964e-08, reward=-1.05e-05, cash_bias=0.1691, portfolio_value=1.003, market_value=1.015, episode=4015\n", + "loss=6.258e-08, reward=2.37e-06, cash_bias=0.139, portfolio_value=1.006, market_value=1.023, episode=4031\n", + "loss=6.792e-08, reward=1.25e-05, cash_bias=0.1784, portfolio_value=1.101, market_value=1.056, episode=4047\n", + "loss=5.492e-08, reward=-1.606e-05, cash_bias=0.1424, portfolio_value=1.002, market_value=1.031, episode=4063\n", + "loss=4.643e-08, reward=3.883e-06, cash_bias=0.06995, portfolio_value=1.031, market_value=1.04, episode=4079\n", + "loss=6.561e-08, reward=9.094e-06, cash_bias=0.2161, portfolio_value=1.056, market_value=1.035, episode=4095\n", + "loss=9.44e-08, reward=-7.065e-06, cash_bias=0.211, portfolio_value=1.009, market_value=1.016, episode=4111\n", + "loss=7.76e-08, reward=1.845e-05, cash_bias=0.2511, portfolio_value=0.9997, market_value=1.005, episode=4127\n", + "loss=8.192e-08, reward=-1.051e-05, cash_bias=0.1257, portfolio_value=1.041, market_value=1.039, episode=4143\n", + "loss=9.571e-08, reward=-8.681e-06, cash_bias=0.1747, portfolio_value=1.026, market_value=1.018, episode=4159\n", + "loss=7.195e-08, reward=1.831e-06, cash_bias=0.1981, portfolio_value=1.033, market_value=1.034, episode=4175\n", + "loss=7.049e-08, reward=2.413e-05, cash_bias=0.277, portfolio_value=1.05, market_value=1.047, episode=4191\n", + "loss=7.244e-08, reward=1.027e-05, cash_bias=0.1616, portfolio_value=1.031, market_value=1.023, episode=4207\n", + "loss=8.402e-08, reward=2.429e-05, cash_bias=0.09402, portfolio_value=1.001, market_value=1.005, episode=4223\n", + "loss=8.6e-08, reward=3.66e-06, cash_bias=0.3477, portfolio_value=1.03, market_value=1.038, episode=4239\n", + "loss=8.24e-08, reward=-4.325e-06, cash_bias=0.09018, portfolio_value=1.033, market_value=1.04, episode=4255\n", + "loss=5.256e-08, reward=8.165e-06, cash_bias=0.1349, portfolio_value=1.053, market_value=1.044, episode=4271\n", + "loss=5.731e-08, reward=-1.079e-05, cash_bias=0.1323, portfolio_value=0.9971, market_value=1.019, episode=4287\n", + "loss=5.45e-08, reward=2.111e-05, cash_bias=0.2006, portfolio_value=1.028, market_value=1.032, episode=4303\n", + "loss=8.468e-08, reward=-1.778e-05, cash_bias=0.2384, portfolio_value=1.02, market_value=1.02, episode=4319\n", + "loss=8.395e-08, reward=1.263e-05, cash_bias=0.1316, portfolio_value=1.037, market_value=1.05, episode=4335\n", + "loss=6.421e-08, reward=-5.325e-06, cash_bias=0.2455, portfolio_value=1.053, market_value=1.042, episode=4351\n", + "loss=5.281e-08, reward=-7.752e-06, cash_bias=0.1391, portfolio_value=1.022, market_value=1.024, episode=4367\n", + "loss=3.893e-08, reward=8.168e-06, cash_bias=0.1557, portfolio_value=1.014, market_value=1.028, episode=4383\n", + "loss=3.067e-08, reward=4.752e-06, cash_bias=0.1645, portfolio_value=1.024, market_value=1.03, episode=4399\n", + "loss=2.391e-08, reward=4.585e-06, cash_bias=0.1665, portfolio_value=1.016, market_value=1.032, episode=4415\n", + "loss=2.827e-08, reward=-1.598e-05, cash_bias=0.2491, portfolio_value=1.039, market_value=1.034, episode=4431\n", + "loss=2.545e-08, reward=-1.096e-05, cash_bias=0.1026, portfolio_value=1.038, market_value=1.053, episode=4447\n", + "loss=2.775e-08, reward=3.024e-05, cash_bias=0.1768, portfolio_value=1.038, market_value=1.065, episode=4463\n", + "loss=2.164e-08, reward=-3.932e-06, cash_bias=0.1688, portfolio_value=1.043, market_value=1.032, episode=4479\n", + "loss=2.582e-08, reward=-2.332e-05, cash_bias=0.1953, portfolio_value=1.011, market_value=1.035, episode=4495\n", + "loss=2.635e-08, reward=1.668e-06, cash_bias=0.1755, portfolio_value=1.041, market_value=1.042, episode=4511\n", + "loss=1.713e-08, reward=-5.439e-06, cash_bias=0.1666, portfolio_value=1.012, market_value=1.027, episode=4527\n", + "loss=2.119e-08, reward=2.399e-05, cash_bias=0.1207, portfolio_value=1.053, market_value=1.047, episode=4543\n", + "loss=1.654e-08, reward=2.628e-06, cash_bias=0.2441, portfolio_value=1.007, market_value=1.004, episode=4559\n", + "loss=1.253e-08, reward=-2.83e-06, cash_bias=0.215, portfolio_value=1.07, market_value=1.054, episode=4575\n", + "loss=1.395e-08, reward=-2.829e-05, cash_bias=0.1807, portfolio_value=1.044, market_value=1.034, episode=4591\n", + "loss=1.041e-08, reward=-1.409e-05, cash_bias=0.2932, portfolio_value=1.075, market_value=1.058, episode=4607\n", + "loss=2.609e-08, reward=-1.523e-06, cash_bias=0.1565, portfolio_value=1.028, market_value=1.06, episode=4623\n", + "loss=9.596e-09, reward=-3.589e-06, cash_bias=0.1931, portfolio_value=1.021, market_value=1.051, episode=4639\n", + "loss=1.42e-08, reward=1.071e-05, cash_bias=0.1099, portfolio_value=1.045, market_value=1.033, episode=4655\n", + "loss=2.052e-08, reward=1.894e-05, cash_bias=0.177, portfolio_value=1.063, market_value=1.045, episode=4671\n", + "loss=2.142e-08, reward=-8.391e-06, cash_bias=0.1956, portfolio_value=1.01, market_value=1.029, episode=4687\n", + "loss=2.362e-08, reward=3.649e-06, cash_bias=0.1625, portfolio_value=1.019, market_value=1.041, episode=4703\n", + "loss=2.544e-08, reward=8.524e-07, cash_bias=0.2077, portfolio_value=1.027, market_value=1.049, episode=4719\n", + "loss=2.212e-08, reward=8.447e-06, cash_bias=0.1065, portfolio_value=0.9999, market_value=1.021, episode=4735\n", + "loss=2.252e-08, reward=2.794e-06, cash_bias=0.2008, portfolio_value=1.038, market_value=1.037, episode=4751\n", + "loss=2.107e-08, reward=5.517e-06, cash_bias=0.1673, portfolio_value=1.01, market_value=1.046, episode=4767\n", + "loss=1.776e-08, reward=7.377e-06, cash_bias=0.2079, portfolio_value=1.039, market_value=1.037, episode=4783\n", + "loss=1.099e-08, reward=9.557e-08, cash_bias=0.1381, portfolio_value=1.008, market_value=1.037, episode=4799\n", + "loss=1.026e-08, reward=4.644e-06, cash_bias=0.1331, portfolio_value=1.008, market_value=1.021, episode=4815\n", + "loss=1.369e-08, reward=7.324e-06, cash_bias=0.2144, portfolio_value=1.048, market_value=1.03, episode=4831\n", + "loss=1.287e-08, reward=2.037e-05, cash_bias=0.2525, portfolio_value=1.056, market_value=1.052, episode=4847\n", + "loss=1.94e-08, reward=6.089e-06, cash_bias=0.1383, portfolio_value=1.036, market_value=1.05, episode=4863\n", + "loss=3.135e-07, reward=7.383e-06, cash_bias=0.1705, portfolio_value=1.018, market_value=1.02, episode=4879\n", + "loss=2.707e-08, reward=1.036e-05, cash_bias=0.1018, portfolio_value=1.041, market_value=1.031, episode=4895\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=3.364e-08, reward=-5.71e-06, cash_bias=0.2188, portfolio_value=1.024, market_value=1.017, episode=4911\n", + "loss=2.995e-08, reward=4.668e-06, cash_bias=0.07444, portfolio_value=1.019, market_value=1.027, episode=4927\n", + "loss=2.218e-08, reward=2.545e-05, cash_bias=0.2557, portfolio_value=1.03, market_value=1.02, episode=4943\n", + "loss=2.243e-08, reward=-1.465e-05, cash_bias=0.2224, portfolio_value=1.011, market_value=1.006, episode=4959\n", + "loss=1.441e-08, reward=3.194e-05, cash_bias=0.2331, portfolio_value=1.037, market_value=1.051, episode=4975\n", + "loss=1.19e-08, reward=1.562e-05, cash_bias=0.1996, portfolio_value=0.9984, market_value=1.012, episode=4991\n", + "loss=1.51e-08, reward=-1.2e-05, cash_bias=0.1422, portfolio_value=1.066, market_value=1.048, episode=5007\n", + "loss=1.617e-08, reward=3.002e-05, cash_bias=0.2592, portfolio_value=1.037, market_value=1.047, episode=5023\n", + "loss=1.147e-08, reward=5.251e-06, cash_bias=0.09229, portfolio_value=1.046, market_value=1.031, episode=5039\n", + "loss=1.665e-08, reward=-6.466e-06, cash_bias=0.1266, portfolio_value=1.004, market_value=1.033, episode=5055\n", + "loss=1.269e-08, reward=2.021e-05, cash_bias=0.1802, portfolio_value=1.06, market_value=1.049, episode=5071\n", + "loss=1.541e-08, reward=2.572e-05, cash_bias=0.1451, portfolio_value=1.034, market_value=1.031, episode=5087\n", + "loss=1.55e-08, reward=1.503e-05, cash_bias=0.1484, portfolio_value=1.043, market_value=1.031, episode=5103\n", + "loss=1.063e-08, reward=2.351e-05, cash_bias=0.2179, portfolio_value=1.036, market_value=1.038, episode=5119\n", + "loss=1.236e-08, reward=1.048e-05, cash_bias=0.2128, portfolio_value=1.049, market_value=1.047, episode=5135\n", + "loss=3.497e-08, reward=1.227e-05, cash_bias=0.1941, portfolio_value=1.052, market_value=1.065, episode=5151\n", + "loss=3.586e-08, reward=1.806e-05, cash_bias=0.1734, portfolio_value=1.018, market_value=1.036, episode=5167\n", + "loss=5.889e-08, reward=-3.957e-06, cash_bias=0.1997, portfolio_value=1.077, market_value=1.054, episode=5183\n", + "loss=8.682e-08, reward=2.375e-05, cash_bias=0.02048, portfolio_value=0.9794, market_value=1.005, episode=5199\n", + "loss=7.757e-08, reward=1.389e-05, cash_bias=0.3183, portfolio_value=0.9988, market_value=1.027, episode=5215\n", + "loss=1.137e-07, reward=-1.239e-05, cash_bias=0.0924, portfolio_value=1.022, market_value=1.035, episode=5231\n", + "loss=1.986e-07, reward=4.71e-06, cash_bias=0.2201, portfolio_value=1.025, market_value=1.048, episode=5247\n", + "loss=2.877e-07, reward=-2.677e-06, cash_bias=0.1539, portfolio_value=1.018, market_value=1.025, episode=5263\n", + "loss=2.838e-07, reward=-3.801e-06, cash_bias=0.1828, portfolio_value=1.056, market_value=1.058, episode=5279\n", + "loss=2.893e-07, reward=1.072e-05, cash_bias=0.1678, portfolio_value=1.01, market_value=1.021, episode=5295\n", + "loss=2.803e-07, reward=-2.721e-06, cash_bias=0.1066, portfolio_value=1.032, market_value=1.036, episode=5311\n", + "loss=2.77e-07, reward=-3.28e-05, cash_bias=0.1941, portfolio_value=1.031, market_value=1.04, episode=5327\n", + "loss=3.071e-07, reward=-3.541e-05, cash_bias=0.1679, portfolio_value=1.02, market_value=1.025, episode=5343\n", + "loss=3.173e-07, reward=-2.282e-06, cash_bias=0.123, portfolio_value=1.059, market_value=1.081, episode=5359\n", + "loss=3.073e-07, reward=-8.825e-06, cash_bias=0.2186, portfolio_value=1.033, market_value=1.062, episode=5375\n", + "loss=3.793e-07, reward=-3.782e-06, cash_bias=0.1005, portfolio_value=1.043, market_value=1.049, episode=5391\n", + "loss=3.619e-07, reward=8.874e-06, cash_bias=0.144, portfolio_value=1.035, market_value=1.038, episode=5407\n", + "loss=3.866e-07, reward=-6.232e-06, cash_bias=0.1519, portfolio_value=1.006, market_value=1.012, episode=5423\n", + "loss=3.907e-07, reward=1.009e-05, cash_bias=0.2202, portfolio_value=1.043, market_value=1.041, episode=5439\n", + "loss=7.077e-07, reward=4.521e-06, cash_bias=0.1728, portfolio_value=1.068, market_value=1.058, episode=5455\n", + "loss=1.944e-06, reward=-2.868e-06, cash_bias=0.1164, portfolio_value=1.017, market_value=1.035, episode=5471\n", + "loss=4.946e-06, reward=1.404e-05, cash_bias=0.2059, portfolio_value=1.024, market_value=1.029, episode=5487\n", + "loss=6.598e-06, reward=-7.398e-06, cash_bias=0.1552, portfolio_value=1.076, market_value=1.053, episode=5503\n", + "loss=8.949e-06, reward=-1.003e-05, cash_bias=0.08738, portfolio_value= 1, market_value=1.033, episode=5519\n", + "loss=1.696e-05, reward=9.563e-06, cash_bias=0.1643, portfolio_value=1.012, market_value=1.022, episode=5535\n", + "loss=1.416e-05, reward=-2.753e-05, cash_bias=0.1625, portfolio_value=0.9994, market_value=1.011, episode=5551\n", + "loss=5.898e-06, reward=9.399e-06, cash_bias=0.1078, portfolio_value=1.046, market_value=1.045, episode=5567\n", + "loss=2.31e-06, reward=-1.973e-06, cash_bias=0.15, portfolio_value=1.034, market_value=1.027, episode=5583\n", + "loss=3.934e-06, reward=5.515e-06, cash_bias=0.2093, portfolio_value=1.054, market_value=1.038, episode=5599\n", + "loss=1.469e-05, reward=5.89e-07, cash_bias=0.1331, portfolio_value=1.017, market_value=1.034, episode=5615\n", + "loss=2.134e-05, reward=-8.087e-06, cash_bias=0.1253, portfolio_value=1.072, market_value=1.038, episode=5631\n", + "loss=2.548e-05, reward=-1.316e-05, cash_bias=0.1348, portfolio_value=1.017, market_value=1.019, episode=5647\n", + "loss=6.698e-06, reward=7.824e-07, cash_bias=0.1683, portfolio_value=1.019, market_value=1.037, episode=5663\n", + "loss=6.121e-07, reward=-5.239e-07, cash_bias=0.1516, portfolio_value=1.009, market_value=1.037, episode=5679\n", + "loss=3.622e-07, reward=1.37e-05, cash_bias=0.1306, portfolio_value=1.046, market_value=1.041, episode=5695\n", + "loss=9.701e-08, reward=-7.639e-06, cash_bias=0.1432, portfolio_value=1.021, market_value=1.016, episode=5711\n", + "loss=2.464e-07, reward=1.722e-05, cash_bias=0.07769, portfolio_value=1.068, market_value=1.065, episode=5727\n", + "loss=2.596e-07, reward=-6.27e-06, cash_bias=0.2099, portfolio_value=1.004, market_value=1.033, episode=5743\n", + "loss=1.807e-07, reward=-1.708e-06, cash_bias=0.1637, portfolio_value=1.042, market_value=1.035, episode=5759\n", + "loss=5.197e-08, reward=3.357e-06, cash_bias=0.1082, portfolio_value=1.03, market_value=1.034, episode=5775\n", + "loss=3.659e-08, reward=1.36e-06, cash_bias=0.1682, portfolio_value=1.06, market_value=1.052, episode=5791\n", + "loss=5.946e-08, reward=3.091e-06, cash_bias=0.197, portfolio_value=1.05, market_value=1.022, episode=5807\n", + "loss=7.123e-08, reward=1.368e-06, cash_bias=0.1216, portfolio_value=1.076, market_value=1.06, episode=5823\n", + "loss=1.23e-07, reward=-2.824e-07, cash_bias=0.09016, portfolio_value=1.009, market_value=1.036, episode=5839\n", + "loss=1.85e-07, reward=1.941e-05, cash_bias=0.1749, portfolio_value=1.049, market_value=1.02, episode=5855\n", + "loss=1.786e-07, reward=-6.442e-06, cash_bias=0.111, portfolio_value=1.048, market_value=1.039, episode=5871\n", + "loss=3.079e-07, reward=-1.379e-05, cash_bias=0.177, portfolio_value=1.078, market_value=1.044, episode=5887\n", + "loss=3.486e-07, reward=-1.247e-05, cash_bias=0.161, portfolio_value=1.01, market_value=1.004, episode=5903\n", + "loss=3.245e-07, reward=5.185e-06, cash_bias=0.2053, portfolio_value=1.038, market_value=1.054, episode=5919\n", + "loss=1.508e-07, reward=4.148e-05, cash_bias=0.1665, portfolio_value=1.008, market_value=1.033, episode=5935\n", + "loss=9.236e-08, reward=1.698e-05, cash_bias=0.1251, portfolio_value=1.063, market_value=1.046, episode=5951\n", + "loss=5.987e-08, reward=2.738e-05, cash_bias=0.09384, portfolio_value=1.011, market_value=1.031, episode=5967\n", + "loss=1.722e-07, reward=7.229e-06, cash_bias=0.2685, portfolio_value=1.05, market_value=1.035, episode=5983\n", + "loss=4.39e-07, reward=-2.132e-05, cash_bias=0.1999, portfolio_value=1.048, market_value=1.049, episode=5999\n", + "loss=6.738e-07, reward=2.886e-05, cash_bias=0.2272, portfolio_value=1.027, market_value=1.016, episode=6015\n", + "loss=1.359e-06, reward=1.078e-05, cash_bias=0.1478, portfolio_value=1.005, market_value=1.008, episode=6031\n", + "loss=2.18e-06, reward=-7.091e-06, cash_bias=0.1902, portfolio_value=1.051, market_value=1.046, episode=6047\n", + "loss=8.785e-06, reward=6.955e-06, cash_bias=0.1875, portfolio_value=1.07, market_value=1.055, episode=6063\n", + "loss=1.392e-05, reward=1.078e-05, cash_bias=0.1776, portfolio_value=1.022, market_value=1.037, episode=6079\n", + "loss=6.386e-06, reward=1.945e-06, cash_bias=0.1715, portfolio_value=1.053, market_value=1.045, episode=6095\n", + "loss=3.558e-07, reward=1.722e-06, cash_bias=0.1929, portfolio_value=1.027, market_value=1.047, episode=6111\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=3.779e-07, reward=2.66e-05, cash_bias=0.1762, portfolio_value=1.033, market_value=1.046, episode=6127\n", + "loss=3.835e-07, reward=2.77e-05, cash_bias=0.1136, portfolio_value=1.034, market_value=1.01, episode=6143\n", + "loss=2.241e-07, reward=-2.945e-06, cash_bias=0.1384, portfolio_value=0.9984, market_value=0.9992, episode=6159\n", + "loss=2.752e-07, reward=-1.832e-05, cash_bias=0.1317, portfolio_value=1.046, market_value=1.019, episode=6175\n", + "loss=2.633e-07, reward=3.253e-06, cash_bias=0.2022, portfolio_value=0.993, market_value=1.015, episode=6191\n", + "loss=2.45e-07, reward=-1.047e-05, cash_bias=0.3177, portfolio_value=1.09, market_value=1.057, episode=6207\n", + "loss=1.262e-07, reward=-6.817e-07, cash_bias=0.1428, portfolio_value=1.037, market_value=1.037, episode=6223\n", + "loss=3.587e-08, reward=-3.061e-06, cash_bias=0.1896, portfolio_value=1.043, market_value=0.9946, episode=6239\n", + "loss=4.273e-08, reward=-1.718e-06, cash_bias=0.1058, portfolio_value=1.075, market_value=1.052, episode=6255\n", + "loss=1.106e-07, reward=4.558e-06, cash_bias=0.2447, portfolio_value=1.056, market_value=1.04, episode=6271\n", + "loss=1.381e-07, reward=1.402e-05, cash_bias=0.1463, portfolio_value=1.051, market_value=1.05, episode=6287\n", + "loss=1.788e-07, reward=3.681e-05, cash_bias=0.1911, portfolio_value=1.058, market_value=1.042, episode=6303\n", + "loss=1.293e-07, reward=1.022e-05, cash_bias=0.206, portfolio_value=1.036, market_value=1.023, episode=6319\n", + "loss=3.53e-08, reward=6.44e-06, cash_bias=0.2762, portfolio_value=1.045, market_value=1.052, episode=6335\n", + "loss=2.039e-08, reward=-1.22e-05, cash_bias=0.1517, portfolio_value=0.9893, market_value=0.9794, episode=6351\n", + "loss=1.523e-08, reward=2.158e-05, cash_bias=0.2026, portfolio_value=1.063, market_value=1.056, episode=6367\n", + "loss=2.956e-08, reward=9.727e-06, cash_bias=0.1634, portfolio_value=1.027, market_value=1.023, episode=6383\n", + "loss=4.104e-08, reward=1.7e-05, cash_bias=0.0822, portfolio_value= 1, market_value=0.9955, episode=6399\n", + "loss=3.183e-08, reward=-1.559e-05, cash_bias=0.1395, portfolio_value=1.041, market_value=1.042, episode=6415\n", + "loss=6.037e-08, reward=-1.888e-05, cash_bias=0.06512, portfolio_value=1.04, market_value=1.017, episode=6431\n", + "loss=5.924e-08, reward=-2.536e-06, cash_bias=0.07191, portfolio_value=1.003, market_value=1.007, episode=6447\n", + "loss=4.535e-08, reward=4.884e-07, cash_bias=0.1746, portfolio_value=1.013, market_value=1.006, episode=6463\n", + "loss=3.402e-08, reward=-3.61e-07, cash_bias=0.1134, portfolio_value=1.033, market_value=1.015, episode=6479\n", + "loss=3.966e-08, reward=2.895e-05, cash_bias=0.1887, portfolio_value=1.031, market_value=1.045, episode=6495\n", + "loss=4.863e-08, reward=8.029e-06, cash_bias=0.1267, portfolio_value=1.029, market_value=1.039, episode=6511\n", + "loss=8.115e-08, reward=1.724e-05, cash_bias=0.2086, portfolio_value=1.06, market_value=1.038, episode=6527\n", + "loss=1.195e-07, reward=9.45e-06, cash_bias=0.1613, portfolio_value=1.039, market_value=1.031, episode=6543\n", + "loss=1.27e-07, reward=-2.082e-06, cash_bias=0.1206, portfolio_value=1.038, market_value=1.056, episode=6559\n", + "loss=6.81e-08, reward=1.546e-06, cash_bias=0.1602, portfolio_value=1.032, market_value=1.017, episode=6575\n", + "loss=5.794e-08, reward=-1.022e-05, cash_bias=0.1696, portfolio_value=1.052, market_value=1.048, episode=6591\n", + "loss=5.062e-08, reward=-1.42e-05, cash_bias=0.09076, portfolio_value=1.037, market_value=1.04, episode=6607\n", + "loss=2.541e-08, reward=1.064e-06, cash_bias=0.1644, portfolio_value=1.013, market_value=1.017, episode=6623\n", + "loss=2.536e-08, reward=3.663e-06, cash_bias=0.1503, portfolio_value=1.023, market_value=1.005, episode=6639\n", + "loss=2.714e-08, reward=-1.336e-05, cash_bias=0.2786, portfolio_value=1.001, market_value=1.02, episode=6655\n", + "loss=4.147e-08, reward=-4.171e-06, cash_bias=0.2387, portfolio_value=1.053, market_value=1.052, episode=6671\n", + "loss=6.932e-08, reward=9.328e-06, cash_bias=0.1001, portfolio_value=1.073, market_value=1.055, episode=6687\n", + "loss=7.328e-08, reward=-2.719e-06, cash_bias=0.1181, portfolio_value=1.02, market_value=1.017, episode=6703\n", + "loss=2.284e-08, reward=4.319e-06, cash_bias=0.2917, portfolio_value=1.027, market_value=1.032, episode=6719\n", + "loss=1.849e-08, reward=-8.377e-07, cash_bias=0.1857, portfolio_value=1.009, market_value=1.026, episode=6735\n", + "loss=2.62e-08, reward=1.737e-05, cash_bias=0.1423, portfolio_value=1.042, market_value=1.018, episode=6751\n", + "loss=1.337e-08, reward=-2.548e-06, cash_bias=0.1305, portfolio_value=1.015, market_value=1.011, episode=6767\n", + "loss=2.319e-08, reward=-9.948e-07, cash_bias=0.2417, portfolio_value=1.027, market_value=1.021, episode=6783\n", + "loss=2.453e-08, reward=-8.083e-06, cash_bias=0.07584, portfolio_value=1.045, market_value=1.033, episode=6799\n", + "loss=1.697e-08, reward=3.874e-05, cash_bias=0.2191, portfolio_value=1.083, market_value=1.06, episode=6815\n", + "loss=2.369e-08, reward=4.752e-06, cash_bias=0.1403, portfolio_value=1.033, market_value=1.048, episode=6831\n", + "loss=4.483e-08, reward=-9.241e-06, cash_bias=0.1163, portfolio_value=1.022, market_value=1.023, episode=6847\n", + "loss=4.991e-08, reward=-1.003e-05, cash_bias=0.0758, portfolio_value=1.045, market_value=1.041, episode=6863\n", + "loss=6.456e-08, reward=-5.918e-08, cash_bias=0.1942, portfolio_value=1.028, market_value=1.036, episode=6879\n", + "loss=9.556e-08, reward=1.133e-05, cash_bias=0.1809, portfolio_value= 1, market_value=1.007, episode=6895\n", + "loss=1.974e-07, reward=3.293e-06, cash_bias=0.2282, portfolio_value=1.024, market_value=1.049, episode=6911\n", + "loss=1.787e-07, reward=1.724e-05, cash_bias=0.1363, portfolio_value=1.014, market_value=1.029, episode=6927\n", + "loss=2.393e-07, reward=-2.194e-05, cash_bias=0.1121, portfolio_value=1.029, market_value=1.026, episode=6943\n", + "loss=4.65e-07, reward=-7.662e-06, cash_bias=0.2362, portfolio_value=1.012, market_value=0.9921, episode=6959\n", + "loss=2.842e-07, reward=8.324e-07, cash_bias=0.109, portfolio_value=1.043, market_value=1.042, episode=6975\n", + "loss=1.437e-07, reward=-1.116e-05, cash_bias=0.1839, portfolio_value=1.048, market_value=1.047, episode=6991\n", + "loss=3.032e-08, reward=2.46e-05, cash_bias=0.1867, portfolio_value=1.043, market_value=1.041, episode=7007\n", + "loss=3.052e-08, reward=-1.224e-05, cash_bias=0.2157, portfolio_value=1.04, market_value=1.068, episode=7023\n", + "loss=2.38e-08, reward=-5.481e-06, cash_bias=0.1736, portfolio_value=1.054, market_value=1.046, episode=7039\n", + "loss=1.615e-08, reward=1.131e-05, cash_bias=0.2422, portfolio_value=0.9962, market_value=1.028, episode=7055\n", + "loss=5.344e-08, reward=4.48e-06, cash_bias=0.1743, portfolio_value=1.024, market_value=1.021, episode=7071\n", + "loss=6.306e-08, reward=-1.419e-05, cash_bias=0.1605, portfolio_value=1.013, market_value=1.02, episode=7087\n", + "loss=1.793e-08, reward=5.204e-06, cash_bias=0.2202, portfolio_value=1.035, market_value=1.035, episode=7103\n", + "loss=1.879e-08, reward=3.882e-06, cash_bias=0.1667, portfolio_value=1.044, market_value=1.018, episode=7119\n", + "loss=4.881e-08, reward=4.592e-06, cash_bias=0.123, portfolio_value=1.005, market_value=1.021, episode=7135\n", + "loss=5.283e-08, reward=-1.076e-05, cash_bias=0.1412, portfolio_value=0.982, market_value=0.9882, episode=7151\n", + "loss=3.72e-08, reward=-6.74e-06, cash_bias=0.1343, portfolio_value=1.051, market_value=1.041, episode=7167\n", + "loss=5.27e-08, reward=3.582e-07, cash_bias=0.1502, portfolio_value=1.008, market_value=1.013, episode=7183\n", + "loss=5.106e-08, reward=8.037e-06, cash_bias=0.2123, portfolio_value=1.037, market_value=1.014, episode=7199\n", + "loss=1.624e-08, reward=9.78e-07, cash_bias=0.1268, portfolio_value=1.019, market_value=1.028, episode=7215\n", + "loss=1.478e-08, reward=5.137e-06, cash_bias=0.2764, portfolio_value=1.043, market_value=1.049, episode=7231\n", + "loss=2.647e-08, reward=2.004e-05, cash_bias=0.1207, portfolio_value=1.069, market_value=1.063, episode=7247\n", + "loss=3.307e-08, reward=-1.206e-05, cash_bias=0.1484, portfolio_value=1.047, market_value=1.039, episode=7263\n", + "loss=4.832e-08, reward=-5.195e-06, cash_bias=0.1676, portfolio_value=1.048, market_value=1.046, episode=7279\n", + "loss=6.117e-08, reward=-1.337e-05, cash_bias=0.1605, portfolio_value=1.023, market_value=1.045, episode=7295\n", + "loss=5.116e-08, reward=1.786e-05, cash_bias=0.1473, portfolio_value=1.078, market_value=1.052, episode=7311\n", + "loss=2.454e-08, reward=1.226e-05, cash_bias=0.1563, portfolio_value=1.024, market_value=1.032, episode=7327\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=3.903e-08, reward=-5.161e-06, cash_bias=0.1298, portfolio_value=1.047, market_value=1.026, episode=7343\n", + "loss=7.84e-08, reward=-3.861e-06, cash_bias=0.1686, portfolio_value=1.057, market_value=1.034, episode=7359\n", + "loss=6.474e-08, reward=-3.305e-05, cash_bias=0.06862, portfolio_value=1.034, market_value=1.021, episode=7375\n", + "loss=7.829e-08, reward=-1.154e-05, cash_bias=0.1207, portfolio_value=1.016, market_value=1.029, episode=7391\n", + "loss=6.964e-08, reward=-8.415e-06, cash_bias=0.06926, portfolio_value=1.055, market_value=1.061, episode=7407\n", + "loss=1.26e-07, reward=9.376e-06, cash_bias=0.1368, portfolio_value=1.035, market_value=1.032, episode=7423\n", + "loss=2.348e-07, reward=-5.839e-06, cash_bias=0.1605, portfolio_value=1.041, market_value=1.045, episode=7439\n", + "loss=5.16e-08, reward=-4.438e-06, cash_bias=0.233, portfolio_value=1.098, market_value=1.075, episode=7455\n", + "loss=2.327e-08, reward=-8.938e-06, cash_bias=0.2203, portfolio_value=1.023, market_value=1.026, episode=7471\n", + "loss=1.058e-07, reward=9.525e-06, cash_bias=0.145, portfolio_value=1.017, market_value=1.039, episode=7487\n", + "loss=3.126e-07, reward=6.301e-06, cash_bias=0.2103, portfolio_value=1.011, market_value=1.021, episode=7503\n", + "loss=4.99e-07, reward=1.316e-05, cash_bias=0.1791, portfolio_value=1.059, market_value=1.041, episode=7519\n", + "loss=4.271e-07, reward=1.849e-06, cash_bias=0.1006, portfolio_value=1.054, market_value=1.042, episode=7535\n", + "loss=6.071e-08, reward=-7.592e-06, cash_bias=0.1877, portfolio_value=1.04, market_value=1.04, episode=7551\n", + "loss=1.675e-07, reward=1.704e-05, cash_bias=0.2022, portfolio_value=1.033, market_value=1.019, episode=7567\n", + "loss=7.542e-08, reward=3.403e-06, cash_bias=0.1434, portfolio_value=1.045, market_value=1.037, episode=7583\n", + "loss=4.076e-08, reward=-1.077e-06, cash_bias=0.2273, portfolio_value=1.054, market_value=1.04, episode=7599\n", + "loss=2.994e-08, reward=2.467e-07, cash_bias=0.1188, portfolio_value=1.068, market_value=1.041, episode=7615\n", + "loss=1.692e-08, reward=-7.83e-06, cash_bias=0.1962, portfolio_value=1.015, market_value=1.037, episode=7631\n", + "loss=1.562e-08, reward=1.779e-05, cash_bias=0.1438, portfolio_value=1.073, market_value=1.07, episode=7647\n", + "loss=1.686e-08, reward=-1.346e-05, cash_bias=0.1862, portfolio_value=1.064, market_value=1.07, episode=7663\n", + "loss=1.951e-08, reward=-3.586e-06, cash_bias=0.1603, portfolio_value=1.032, market_value=1.019, episode=7679\n", + "loss=3.965e-08, reward=-1.189e-05, cash_bias=0.2359, portfolio_value=1.032, market_value=1.023, episode=7695\n", + "loss=4.539e-08, reward=1.08e-05, cash_bias=0.1934, portfolio_value=1.051, market_value=1.027, episode=7711\n", + "loss=2.426e-08, reward=1.577e-05, cash_bias=0.04825, portfolio_value=1.049, market_value=1.042, episode=7727\n", + "loss=3.593e-08, reward=1.735e-05, cash_bias=0.08502, portfolio_value=1.037, market_value=1.051, episode=7743\n", + "loss=2.564e-08, reward=1.268e-05, cash_bias=0.1795, portfolio_value=0.9987, market_value=1.023, episode=7759\n", + "loss=1.864e-08, reward=-7.057e-06, cash_bias=0.2507, portfolio_value=1.044, market_value=1.036, episode=7775\n", + "loss=2.379e-08, reward=-4.796e-05, cash_bias=0.1569, portfolio_value=1.041, market_value=1.036, episode=7791\n", + "loss=3.695e-08, reward=-7.642e-06, cash_bias=0.1068, portfolio_value=1.064, market_value=1.043, episode=7807\n", + "loss=1.701e-08, reward=2.306e-05, cash_bias=0.1191, portfolio_value=1.052, market_value=1.048, episode=7823\n", + "loss=4.372e-08, reward=2.26e-05, cash_bias=0.04499, portfolio_value=1.023, market_value=1.016, episode=7839\n", + "loss=1.077e-07, reward=-3.188e-05, cash_bias=0.1244, portfolio_value=1.005, market_value=1.011, episode=7855\n", + "loss=5.708e-07, reward=-5.744e-07, cash_bias=0.244, portfolio_value=0.994, market_value=1.015, episode=7871\n", + "loss=7.292e-06, reward=6.648e-07, cash_bias=0.1214, portfolio_value=1.033, market_value=1.018, episode=7887\n", + "loss=1.002e-05, reward=4.077e-05, cash_bias=0.1122, portfolio_value=1.06, market_value=1.034, episode=7903\n", + "loss=8.904e-06, reward=1.691e-05, cash_bias=0.1652, portfolio_value=1.05, market_value=1.056, episode=7919\n", + "loss=1.702e-05, reward=7.994e-06, cash_bias=0.1659, portfolio_value=1.003, market_value=1.04, episode=7935\n", + "loss=0.000136, reward=1.348e-05, cash_bias=0.1375, portfolio_value=1.067, market_value=1.057, episode=7951\n", + "loss=1.643e-05, reward=-9.955e-06, cash_bias=0.1069, portfolio_value=1.024, market_value=1.023, episode=7967\n", + "loss=1.6e-05, reward=1.552e-05, cash_bias=0.1588, portfolio_value=1.051, market_value=1.048, episode=7983\n", + "loss=1.51e-05, reward=-9.144e-06, cash_bias=0.1711, portfolio_value=1.024, market_value=1.047, episode=7999\n", + "loss=7.471e-06, reward=-8.609e-08, cash_bias=0.1637, portfolio_value=1.004, market_value=1.027, episode=8015\n", + "loss=3.515e-06, reward=-3.943e-06, cash_bias=0.2202, portfolio_value=1.012, market_value=1.04, episode=8031\n", + "loss=1.736e-06, reward=4.838e-06, cash_bias=0.1371, portfolio_value=1.007, market_value=1.024, episode=8047\n", + "loss=4.483e-06, reward=3.3e-06, cash_bias=0.1127, portfolio_value=1.015, market_value=1.021, episode=8063\n", + "loss=2.423e-05, reward=2.711e-06, cash_bias=0.09881, portfolio_value=1.038, market_value=1.037, episode=8079\n", + "loss=3.789e-06, reward=1.12e-05, cash_bias=0.2474, portfolio_value=1.041, market_value=1.029, episode=8095\n", + "loss=1.472e-05, reward=2.94e-05, cash_bias=0.1165, portfolio_value=1.05, market_value=1.043, episode=8111\n", + "loss=5.27e-05, reward=5.806e-06, cash_bias=0.1759, portfolio_value=1.036, market_value=1.029, episode=8127\n", + "loss=0.0005916, reward=7.336e-06, cash_bias=0.1367, portfolio_value=1.061, market_value=1.046, episode=8143\n", + "loss=0.0004028, reward=-2.521e-06, cash_bias=0.212, portfolio_value=1.053, market_value=1.05, episode=8159\n", + "loss=0.0001342, reward=2.016e-05, cash_bias=0.08364, portfolio_value=1.029, market_value=1.021, episode=8175\n", + "loss=0.0002304, reward=-4.722e-06, cash_bias=0.1995, portfolio_value=1.009, market_value=1.02, episode=8191\n", + "loss=0.0002852, reward=-5.71e-06, cash_bias=0.09116, portfolio_value=1.011, market_value=1.013, episode=8207\n", + "loss=7.65e-05, reward=1.538e-05, cash_bias=0.1679, portfolio_value=1.025, market_value=1.033, episode=8223\n", + "loss=2.838e-05, reward=1.137e-05, cash_bias=0.1166, portfolio_value=1.034, market_value=1.031, episode=8239\n", + "loss=8.923e-05, reward=-2.153e-05, cash_bias=0.1943, portfolio_value=1.03, market_value=1.009, episode=8255\n", + "loss=0.000644, reward=2.2e-06, cash_bias=0.1565, portfolio_value=1.06, market_value=1.044, episode=8271\n", + "loss=0.00295, reward=1.807e-05, cash_bias=0.1507, portfolio_value=1.048, market_value=1.036, episode=8287\n", + "loss=0.007676, reward=2.87e-06, cash_bias=0.1317, portfolio_value=1.012, market_value=1.027, episode=8303\n", + "loss=0.00168, reward=2.533e-05, cash_bias=0.1988, portfolio_value=1.044, market_value=1.047, episode=8319\n", + "loss=0.00349, reward=-1.572e-05, cash_bias=0.181, portfolio_value=1.048, market_value=1.028, episode=8335\n", + "loss=0.003512, reward=-3.967e-06, cash_bias=0.123, portfolio_value=1.049, market_value=1.034, episode=8351\n", + "loss=0.001234, reward=8.745e-06, cash_bias=0.1914, portfolio_value=1.022, market_value=1.005, episode=8367\n", + "loss=0.002296, reward=-1.326e-05, cash_bias=0.1584, portfolio_value=1.01, market_value=1.008, episode=8383\n", + "loss=0.001389, reward=1.294e-06, cash_bias=0.1581, portfolio_value=1.051, market_value=1.039, episode=8399\n", + "loss=0.0001181, reward=-4.314e-06, cash_bias=0.1358, portfolio_value=1.07, market_value=1.047, episode=8415\n", + "loss=0.0008767, reward=1.971e-05, cash_bias=0.2565, portfolio_value=1.091, market_value=1.057, episode=8431\n", + "loss=0.001621, reward=2.89e-05, cash_bias=0.1264, portfolio_value=1.109, market_value=1.069, episode=8447\n", + "loss=0.0005708, reward=-2.522e-06, cash_bias=0.1425, portfolio_value=1.044, market_value=1.026, episode=8463\n", + "loss=2.583e-05, reward=-3.652e-06, cash_bias=0.1417, portfolio_value=1.031, market_value=1.006, episode=8479\n", + "loss=0.0001429, reward=-4.486e-05, cash_bias=0.07382, portfolio_value=1.018, market_value=1.024, episode=8495\n", + "loss=0.0002293, reward=9.116e-06, cash_bias=0.1981, portfolio_value=1.058, market_value=1.032, episode=8511\n", + "loss=0.0002602, reward=-1.468e-05, cash_bias=0.1324, portfolio_value=1.056, market_value=1.043, episode=8527\n", + "loss=0.0002697, reward=-6.256e-07, cash_bias=0.1367, portfolio_value=1.058, market_value=1.039, episode=8543\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=0.0002716, reward=9.749e-06, cash_bias=0.1508, portfolio_value=1.075, market_value=1.027, episode=8559\n", + "loss=0.0002702, reward=1.781e-06, cash_bias=0.1204, portfolio_value=1.034, market_value=1.03, episode=8575\n", + "loss=0.0002706, reward=-9.949e-06, cash_bias=0.1874, portfolio_value=1.057, market_value=1.031, episode=8591\n", + "loss=0.0002702, reward=2.466e-05, cash_bias=0.07732, portfolio_value=1.071, market_value=1.041, episode=8607\n", + "loss=0.0002685, reward=6.08e-06, cash_bias=0.104, portfolio_value=1.093, market_value=1.054, episode=8623\n", + "loss=0.0002696, reward=-1.121e-06, cash_bias=0.1782, portfolio_value=1.039, market_value=1.018, episode=8639\n", + "loss=0.0002723, reward=-1.218e-05, cash_bias=0.2466, portfolio_value=1.001, market_value=1.006, episode=8655\n", + "loss=0.0002732, reward=1.98e-05, cash_bias=0.2053, portfolio_value=1.047, market_value=1.035, episode=8671\n", + "loss=0.0002711, reward=-2.625e-05, cash_bias=0.11, portfolio_value=1.026, market_value=1.004, episode=8687\n", + "loss=0.0002707, reward=-5.143e-06, cash_bias=0.05164, portfolio_value=1.082, market_value=1.019, episode=8703\n", + "loss=0.0002718, reward=9.407e-07, cash_bias=0.153, portfolio_value=1.022, market_value=1.027, episode=8719\n", + "loss=0.0002714, reward=1.452e-05, cash_bias=0.09532, portfolio_value=1.017, market_value=1.025, episode=8735\n", + "loss=0.0002707, reward=1.066e-05, cash_bias=0.04865, portfolio_value=1.077, market_value=1.049, episode=8751\n", + "loss=0.000271, reward=-1.642e-05, cash_bias=0.1488, portfolio_value=1.067, market_value=1.033, episode=8767\n", + "loss=0.000271, reward=5.288e-06, cash_bias=0.163, portfolio_value=1.04, market_value=1.025, episode=8783\n", + "loss=0.0002708, reward=1.321e-05, cash_bias=0.1944, portfolio_value=1.111, market_value=1.065, episode=8799\n", + "loss=0.0002682, reward=-1.061e-05, cash_bias=0.1425, portfolio_value=1.042, market_value=1.046, episode=8815\n", + "loss=0.0002658, reward=1.754e-05, cash_bias=0.1776, portfolio_value=1.071, market_value=1.041, episode=8831\n", + "loss=0.0002651, reward=-3.655e-06, cash_bias=0.1002, portfolio_value=1.073, market_value=1.039, episode=8847\n", + "loss=0.0002629, reward=1.993e-06, cash_bias=0.1031, portfolio_value=1.135, market_value=1.084, episode=8863\n", + "loss=0.000261, reward=1.952e-05, cash_bias=0.2004, portfolio_value=1.036, market_value=1.029, episode=8879\n", + "loss=0.0002577, reward=-4.388e-06, cash_bias=0.08845, portfolio_value=1.083, market_value=1.032, episode=8895\n", + "loss=0.0002547, reward=-3.537e-06, cash_bias=0.1472, portfolio_value=1.011, market_value=1.006, episode=8911\n", + "loss=0.0002536, reward=1.297e-05, cash_bias=0.136, portfolio_value=1.043, market_value=1.026, episode=8927\n", + "loss=0.0002534, reward=-1.967e-05, cash_bias=0.1304, portfolio_value=1.042, market_value=1.023, episode=8943\n", + "loss=0.0002534, reward=-1.299e-05, cash_bias=0.0913, portfolio_value=1.064, market_value=1.013, episode=8959\n", + "loss=0.0002527, reward=-6.896e-07, cash_bias=0.1612, portfolio_value=1.035, market_value=1.039, episode=8975\n", + "loss=0.0002524, reward=-2.233e-05, cash_bias=0.09031, portfolio_value=1.044, market_value=1.034, episode=8991\n", + "loss=0.0002533, reward=-1.081e-05, cash_bias=0.06648, portfolio_value=1.029, market_value=1.027, episode=9007\n", + "loss=0.0002533, reward=-1.65e-05, cash_bias=0.1145, portfolio_value=1.048, market_value=1.032, episode=9023\n", + "loss=0.0002546, reward=9.623e-06, cash_bias=0.1537, portfolio_value=1.029, market_value=1.016, episode=9039\n", + "loss=0.0002547, reward=-4.006e-06, cash_bias=0.117, portfolio_value=1.09, market_value=1.058, episode=9055\n", + "loss=0.0002562, reward=5.503e-06, cash_bias=0.2192, portfolio_value=1.046, market_value=1.034, episode=9071\n", + "loss=0.0002567, reward=1.71e-06, cash_bias=0.1535, portfolio_value=1.026, market_value=1.041, episode=9087\n", + "loss=0.0002565, reward=-1.396e-05, cash_bias=0.1882, portfolio_value=1.04, market_value=1.013, episode=9103\n", + "loss=0.000254, reward=-1.573e-07, cash_bias=0.08421, portfolio_value=1.038, market_value=1.015, episode=9119\n", + "loss=0.0002534, reward=-1.043e-05, cash_bias=0.03095, portfolio_value=1.048, market_value=1.042, episode=9135\n", + "loss=0.0002522, reward=3.093e-06, cash_bias=0.1209, portfolio_value=1.039, market_value=1.015, episode=9151\n", + "loss=0.0002525, reward=3.247e-05, cash_bias=0.1111, portfolio_value=1.044, market_value=1.033, episode=9167\n", + "loss=0.0002522, reward=2.789e-06, cash_bias=0.08148, portfolio_value=1.026, market_value=1.002, episode=9183\n", + "loss=0.0002508, reward=1.062e-05, cash_bias=0.03994, portfolio_value=1.069, market_value=1.037, episode=9199\n", + "loss=0.0002482, reward=4.153e-06, cash_bias=0.1932, portfolio_value=1.103, market_value=1.059, episode=9215\n", + "loss=0.0002456, reward=-7.527e-06, cash_bias=0.08757, portfolio_value=1.1, market_value=1.065, episode=9231\n", + "loss=0.0002447, reward=-2.064e-05, cash_bias=0.1264, portfolio_value=1.053, market_value=1.03, episode=9247\n", + "loss=0.0002488, reward=5.666e-06, cash_bias=0.2073, portfolio_value=1.06, market_value=1.059, episode=9263\n", + "loss=0.0002516, reward=5.929e-06, cash_bias=0.2653, portfolio_value=1.021, market_value=1.037, episode=9279\n", + "loss=0.0002526, reward=-1.515e-06, cash_bias=0.1511, portfolio_value=1.046, market_value=1.049, episode=9295\n", + "loss=0.0002536, reward=-5.481e-06, cash_bias=0.08473, portfolio_value=1.097, market_value=1.049, episode=9311\n", + "loss=0.0002528, reward=-1.222e-05, cash_bias=0.1531, portfolio_value=1.044, market_value=1.038, episode=9327\n", + "loss=0.0002532, reward=4.565e-06, cash_bias=0.1821, portfolio_value=1.08, market_value=1.017, episode=9343\n", + "loss=0.0002534, reward=3.221e-06, cash_bias=0.08257, portfolio_value=1.056, market_value=1.043, episode=9359\n", + "loss=0.0002551, reward=-7.266e-06, cash_bias=0.1338, portfolio_value=1.089, market_value=1.045, episode=9375\n", + "loss=0.0002572, reward=-2.231e-06, cash_bias=0.1737, portfolio_value=1.039, market_value=1.013, episode=9391\n", + "loss=0.0002587, reward=-4.804e-06, cash_bias=0.1444, portfolio_value=1.064, market_value=1.033, episode=9407\n", + "loss=0.0002585, reward=-8.131e-06, cash_bias=0.2256, portfolio_value=1.069, market_value=1.045, episode=9423\n", + "loss=0.0002591, reward=1.382e-05, cash_bias=0.2433, portfolio_value=1.033, market_value=1.018, episode=9439\n", + "loss=0.000262, reward=3.279e-06, cash_bias=0.1611, portfolio_value=1.053, market_value=1.037, episode=9455\n", + "loss=0.0002656, reward=1.528e-05, cash_bias=0.2017, portfolio_value=1.075, market_value=1.054, episode=9471\n", + "loss=0.0002706, reward=2.2e-05, cash_bias=0.1116, portfolio_value=1.065, market_value=1.044, episode=9487\n", + "loss=0.0002753, reward=1.524e-05, cash_bias=0.125, portfolio_value=1.042, market_value=1.019, episode=9503\n", + "loss=0.0002756, reward=-1.038e-05, cash_bias=0.1717, portfolio_value=1.026, market_value=1.033, episode=9519\n", + "loss=0.0002742, reward=1.28e-06, cash_bias=0.168, portfolio_value=1.062, market_value=1.045, episode=9535\n", + "loss=0.000271, reward=9.149e-06, cash_bias=0.1327, portfolio_value=1.081, market_value=1.054, episode=9551\n", + "loss=0.0002654, reward=4.036e-06, cash_bias=0.123, portfolio_value=1.059, market_value=1.027, episode=9567\n", + "loss=0.0002534, reward=5.675e-06, cash_bias=0.1467, portfolio_value=1.097, market_value=1.074, episode=9583\n", + "loss=0.0002407, reward=7.306e-06, cash_bias=0.1853, portfolio_value=1.045, market_value=1.014, episode=9599\n", + "loss=0.0002236, reward=-2.678e-06, cash_bias=0.1768, portfolio_value=1.08, market_value=1.041, episode=9615\n", + "loss=0.0002035, reward=2.077e-05, cash_bias=0.1458, portfolio_value=1.097, market_value=1.038, episode=9631\n", + "loss=0.0001853, reward=2.998e-06, cash_bias=0.1675, portfolio_value=1.07, market_value=1.054, episode=9647\n", + "loss=0.0001707, reward=1.978e-05, cash_bias=0.1165, portfolio_value=1.086, market_value=1.055, episode=9663\n", + "loss=0.0001604, reward=2.597e-05, cash_bias=0.1086, portfolio_value=1.025, market_value=1.028, episode=9679\n", + "loss=0.0001515, reward=1.445e-05, cash_bias=0.1466, portfolio_value=1.078, market_value=1.042, episode=9695\n", + "loss=0.0001428, reward=-1.849e-05, cash_bias=0.1597, portfolio_value=1.026, market_value=1.032, episode=9711\n", + "loss=0.0001397, reward=-1.851e-05, cash_bias=0.08933, portfolio_value=1.093, market_value=1.056, episode=9727\n", + "loss=0.0001409, reward=5.823e-06, cash_bias=0.1407, portfolio_value=1.067, market_value=1.051, episode=9743\n", + "loss=0.0001356, reward=7.539e-06, cash_bias=0.165, portfolio_value=1.048, market_value=1.043, episode=9759\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=0.0001281, reward=-4.252e-06, cash_bias=0.1277, portfolio_value=1.018, market_value=1.014, episode=9775\n", + "loss=0.000125, reward=-1.608e-05, cash_bias=0.1199, portfolio_value=1.067, market_value=1.037, episode=9791\n", + "loss=0.0001257, reward=2.874e-05, cash_bias=0.1332, portfolio_value=1.049, market_value= 1, episode=9807\n", + "loss=0.0001217, reward=1.343e-05, cash_bias=0.148, portfolio_value=1.018, market_value=1.023, episode=9823\n", + "loss=0.0001228, reward=-3.833e-06, cash_bias=0.1224, portfolio_value=1.072, market_value=1.046, episode=9839\n", + "loss=0.000129, reward=5.995e-06, cash_bias=0.1109, portfolio_value=1.049, market_value=1.028, episode=9855\n", + "loss=0.0001395, reward=-4.553e-06, cash_bias=0.09929, portfolio_value=1.058, market_value=1.039, episode=9871\n", + "loss=0.0001472, reward=6.648e-07, cash_bias=0.1741, portfolio_value=1.049, market_value=1.001, episode=9887\n", + "loss=0.0001427, reward=-5.906e-06, cash_bias=0.09462, portfolio_value=0.9979, market_value=1.002, episode=9903\n", + "loss=0.0001311, reward=-1.427e-05, cash_bias=0.1417, portfolio_value=1.052, market_value=1.041, episode=9919\n", + "loss=0.0001211, reward=-4.364e-06, cash_bias=0.1286, portfolio_value=1.012, market_value=1.019, episode=9935\n", + "loss=0.0001184, reward=-1.53e-05, cash_bias=0.1096, portfolio_value=1.039, market_value=1.02, episode=9951\n", + "loss=0.0001249, reward=3.522e-06, cash_bias=0.0369, portfolio_value=1.055, market_value=1.045, episode=9967\n", + "loss=0.0001329, reward=2.521e-05, cash_bias=0.1189, portfolio_value=1.065, market_value=1.061, episode=9983\n", + "loss=0.000136, reward=2.483e-07, cash_bias=0.06746, portfolio_value=1.038, market_value=1.015, episode=9999\n", + "loss=0.0001351, reward=-9.627e-06, cash_bias=0.04061, portfolio_value=1.047, market_value=1.043, episode=1.002e+04\n", + "loss=0.0001358, reward=-6.01e-06, cash_bias=0.1151, portfolio_value=1.034, market_value=1.035, episode=1.003e+04\n", + "loss=0.0001445, reward=4.947e-06, cash_bias=0.2377, portfolio_value=1.023, market_value=1.019, episode=1.005e+04\n", + "loss=0.0001498, reward=1.212e-05, cash_bias=0.1881, portfolio_value=1.061, market_value=0.9992, episode=1.006e+04\n", + "loss=0.0001527, reward=3.052e-06, cash_bias=0.2608, portfolio_value=1.059, market_value=1.06, episode=1.008e+04\n", + "loss=0.0001505, reward=3.303e-05, cash_bias=0.138, portfolio_value=1.073, market_value=1.076, episode=1.01e+04\n", + "loss=0.0001466, reward=1.159e-07, cash_bias=0.2068, portfolio_value=1.061, market_value=1.042, episode=1.011e+04\n", + "loss=0.0001454, reward=-4.914e-06, cash_bias=0.1424, portfolio_value=1.044, market_value=1.037, episode=1.013e+04\n", + "loss=0.0001438, reward=9.273e-06, cash_bias=0.1666, portfolio_value=1.104, market_value=1.023, episode=1.014e+04\n", + "loss=0.0001399, reward=-3.672e-06, cash_bias=0.1207, portfolio_value=1.023, market_value=1.016, episode=1.016e+04\n", + "loss=0.0001388, reward=2.349e-05, cash_bias=0.05326, portfolio_value=1.081, market_value=1.042, episode=1.018e+04\n", + "loss=0.0001397, reward=1.39e-05, cash_bias=0.1909, portfolio_value=1.051, market_value=1.048, episode=1.019e+04\n", + "loss=0.0001413, reward=9.531e-06, cash_bias=0.1571, portfolio_value=1.071, market_value=1.03, episode=1.021e+04\n", + "loss=0.000145, reward=1.105e-06, cash_bias=0.1727, portfolio_value=1.026, market_value=1.022, episode=1.022e+04\n", + "loss=0.0001502, reward=1.581e-05, cash_bias=0.125, portfolio_value=1.029, market_value=1.039, episode=1.024e+04\n", + "loss=0.0001549, reward=8.573e-06, cash_bias=0.1948, portfolio_value=1.075, market_value=1.027, episode=1.026e+04\n", + "loss=0.0001601, reward=-1.724e-05, cash_bias=0.1017, portfolio_value=1.073, market_value=1.059, episode=1.027e+04\n", + "loss=0.0001666, reward=-4.647e-05, cash_bias=0.1571, portfolio_value=1.072, market_value=1.035, episode=1.029e+04\n", + "loss=0.0001736, reward=-5.447e-06, cash_bias=0.1153, portfolio_value=1.026, market_value=1.016, episode=1.03e+04\n", + "loss=0.0001817, reward=6.554e-07, cash_bias=0.1401, portfolio_value=1.031, market_value=1.028, episode=1.032e+04\n", + "loss=0.0001882, reward=9.466e-06, cash_bias=0.1479, portfolio_value=1.07, market_value=1.018, episode=1.034e+04\n", + "loss=0.0001946, reward=-1.958e-05, cash_bias=0.08504, portfolio_value=1.054, market_value=1.031, episode=1.035e+04\n", + "loss=0.0002037, reward=1.299e-06, cash_bias=0.1844, portfolio_value=1.062, market_value=1.044, episode=1.037e+04\n", + "loss=0.0002107, reward=-1.116e-06, cash_bias=0.1242, portfolio_value=1.045, market_value=1.032, episode=1.038e+04\n", + "loss=0.0002182, reward=-1.366e-05, cash_bias=0.1213, portfolio_value=1.038, market_value=1.018, episode=1.04e+04\n", + "loss=0.0002266, reward=1.534e-05, cash_bias=0.1097, portfolio_value=1.059, market_value=1.033, episode=1.042e+04\n", + "loss=0.000233, reward=1.038e-05, cash_bias=0.09097, portfolio_value=1.045, market_value=1.039, episode=1.043e+04\n", + "loss=0.0002387, reward=1.056e-05, cash_bias=0.178, portfolio_value=1.051, market_value=1.033, episode=1.045e+04\n", + "loss=0.0002459, reward=4.328e-07, cash_bias=0.1289, portfolio_value=1.044, market_value=1.018, episode=1.046e+04\n", + "loss=0.0002498, reward=1.084e-05, cash_bias=0.1187, portfolio_value=1.017, market_value=0.999, episode=1.048e+04\n", + "loss=0.0002521, reward=5.422e-05, cash_bias=0.09954, portfolio_value=1.072, market_value=1.028, episode=1.05e+04\n", + "loss=0.0002514, reward=8.587e-06, cash_bias=0.08738, portfolio_value=1.065, market_value=1.062, episode=1.051e+04\n", + "loss=0.0002492, reward=2.091e-05, cash_bias=0.15, portfolio_value=1.066, market_value=1.038, episode=1.053e+04\n", + "loss=0.0002454, reward=4.408e-06, cash_bias=0.06882, portfolio_value=1.106, market_value=1.049, episode=1.054e+04\n", + "loss=0.0002321, reward=2.66e-06, cash_bias=0.1312, portfolio_value=1.059, market_value=1.039, episode=1.056e+04\n", + "loss=0.0001952, reward=1.293e-05, cash_bias=0.1507, portfolio_value=1.026, market_value=1.019, episode=1.058e+04\n", + "loss=9.557e-05, reward=-4.121e-06, cash_bias=0.1713, portfolio_value=1.02, market_value=1.021, episode=1.059e+04\n", + "loss=6.914e-06, reward=-1.164e-05, cash_bias=0.04192, portfolio_value=1.032, market_value=0.9933, episode=1.061e+04\n", + "loss=1.368e-05, reward=-2.59e-06, cash_bias=0.04033, portfolio_value=1.083, market_value=1.023, episode=1.062e+04\n", + "loss=3.644e-05, reward=1.135e-05, cash_bias=0.1388, portfolio_value=1.023, market_value=1.008, episode=1.064e+04\n", + "loss=6.002e-05, reward=3.755e-06, cash_bias=0.1233, portfolio_value=1.056, market_value=1.043, episode=1.066e+04\n", + "loss=8.962e-05, reward=-1.721e-05, cash_bias=0.08762, portfolio_value=1.062, market_value=1.018, episode=1.067e+04\n", + "loss=0.0001347, reward=5.917e-06, cash_bias=0.118, portfolio_value=1.046, market_value=1.03, episode=1.069e+04\n", + "loss=0.0001936, reward=1.079e-05, cash_bias=0.139, portfolio_value=1.056, market_value=1.053, episode=1.07e+04\n", + "loss=0.0002109, reward=2.371e-05, cash_bias=0.1769, portfolio_value=0.992, market_value=0.9973, episode=1.072e+04\n", + "loss=0.0001631, reward=-1.159e-05, cash_bias=0.07314, portfolio_value=1.054, market_value=1.026, episode=1.074e+04\n", + "loss=0.0001095, reward=-3.842e-06, cash_bias=0.06879, portfolio_value=1.039, market_value=1.033, episode=1.075e+04\n", + "loss=5.466e-05, reward=-1.007e-06, cash_bias=0.1102, portfolio_value=1.058, market_value=1.036, episode=1.077e+04\n", + "loss=1.895e-05, reward=-1.792e-05, cash_bias=0.1056, portfolio_value=1.047, market_value=1.015, episode=1.078e+04\n", + "loss=9.652e-06, reward=1.649e-05, cash_bias=0.1835, portfolio_value=1.033, market_value=1.02, episode=1.08e+04\n", + "loss=7.64e-06, reward=1.426e-05, cash_bias=0.1131, portfolio_value=1.054, market_value=1.047, episode=1.082e+04\n", + "loss=7.138e-06, reward=6.165e-06, cash_bias=0.07579, portfolio_value=1.056, market_value=1.023, episode=1.083e+04\n", + "loss=6.959e-06, reward=3.625e-06, cash_bias=0.1119, portfolio_value=1.053, market_value=1.027, episode=1.085e+04\n", + "loss=7.043e-06, reward=3.012e-05, cash_bias=0.1275, portfolio_value=1.072, market_value=1.044, episode=1.086e+04\n", + "loss=7.228e-06, reward=-4.617e-06, cash_bias=0.04445, portfolio_value=1.07, market_value=1.044, episode=1.088e+04\n", + "loss=7.498e-06, reward=-1.173e-05, cash_bias=0.1563, portfolio_value=1.092, market_value=1.034, episode=1.09e+04\n", + "loss=7.675e-06, reward=-1.162e-05, cash_bias=0.1251, portfolio_value=1.067, market_value=1.037, episode=1.091e+04\n", + "loss=7.662e-06, reward=-1.505e-05, cash_bias=0.1509, portfolio_value=1.056, market_value=1.033, episode=1.093e+04\n", + "loss=7.542e-06, reward=1.419e-05, cash_bias=0.101, portfolio_value=1.068, market_value=1.023, episode=1.094e+04\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=7.596e-06, reward=-8.442e-05, cash_bias=0.168, portfolio_value=1.036, market_value=1.019, episode=1.096e+04\n", + "loss=7.576e-06, reward=1.155e-05, cash_bias=0.09953, portfolio_value=1.043, market_value=1.026, episode=1.098e+04\n", + "loss=7.402e-06, reward=1.139e-05, cash_bias=0.1416, portfolio_value=1.057, market_value=1.035, episode=1.099e+04\n", + "loss=7.295e-06, reward=-1.263e-05, cash_bias=0.09514, portfolio_value=1.101, market_value=1.051, episode=1.101e+04\n", + "loss=7.112e-06, reward=-1.314e-05, cash_bias=0.09396, portfolio_value=1.081, market_value=1.034, episode=1.102e+04\n", + "loss=6.654e-06, reward=1.633e-05, cash_bias=0.05397, portfolio_value=1.042, market_value=1.034, episode=1.104e+04\n", + "loss=6.775e-06, reward=2.004e-05, cash_bias=0.1456, portfolio_value=1.067, market_value=1.051, episode=1.106e+04\n", + "loss=6.764e-06, reward=-7.112e-06, cash_bias=0.1195, portfolio_value=1.047, market_value=1.021, episode=1.107e+04\n", + "loss=6.702e-06, reward=3.904e-08, cash_bias=0.05377, portfolio_value=1.059, market_value=1.007, episode=1.109e+04\n", + "loss=6.874e-06, reward=1.529e-05, cash_bias=0.1212, portfolio_value=1.058, market_value=1.023, episode=1.11e+04\n", + "loss=6.926e-06, reward=2.224e-05, cash_bias=0.1603, portfolio_value=1.072, market_value=1.043, episode=1.112e+04\n", + "loss=6.901e-06, reward=1.782e-05, cash_bias=0.08408, portfolio_value=1.057, market_value=1.029, episode=1.114e+04\n", + "loss=6.424e-06, reward=2.551e-05, cash_bias=0.09913, portfolio_value=1.038, market_value=1.017, episode=1.115e+04\n", + "loss=6.414e-06, reward=1.961e-05, cash_bias=0.1714, portfolio_value=1.066, market_value=1.028, episode=1.117e+04\n", + "loss=6.469e-06, reward=1.428e-05, cash_bias=0.0974, portfolio_value=1.078, market_value=1.051, episode=1.118e+04\n", + "loss=6.174e-06, reward=-4.079e-06, cash_bias=0.1699, portfolio_value=1.018, market_value=1.004, episode=1.12e+04\n", + "loss=6.209e-06, reward=-1.109e-06, cash_bias=0.113, portfolio_value=1.062, market_value=1.025, episode=1.122e+04\n", + "loss=6.236e-06, reward=-9.98e-06, cash_bias=0.1703, portfolio_value=1.108, market_value=1.054, episode=1.123e+04\n", + "loss=6.108e-06, reward=7.997e-06, cash_bias=0.04572, portfolio_value=1.04, market_value=1.033, episode=1.125e+04\n", + "loss=6.12e-06, reward=-1.415e-05, cash_bias=0.1561, portfolio_value=1.044, market_value=1.043, episode=1.126e+04\n", + "loss=6.091e-06, reward=1.952e-05, cash_bias=0.1403, portfolio_value=1.051, market_value=1.024, episode=1.128e+04\n", + "loss=5.956e-06, reward=-2.175e-05, cash_bias=0.1265, portfolio_value=1.063, market_value=1.02, episode=1.13e+04\n", + "loss=6.1e-06, reward=-4.231e-06, cash_bias=0.1093, portfolio_value=1.085, market_value=1.054, episode=1.131e+04\n", + "loss=6.059e-06, reward=9.711e-06, cash_bias=0.1422, portfolio_value=1.104, market_value=1.068, episode=1.133e+04\n", + "loss=6.061e-06, reward=8.77e-06, cash_bias=0.1419, portfolio_value=1.065, market_value=1.043, episode=1.134e+04\n", + "loss=6.064e-06, reward=-8.16e-06, cash_bias=0.2263, portfolio_value=1.07, market_value=1.046, episode=1.136e+04\n", + "loss=5.941e-06, reward=-4.762e-06, cash_bias=0.1473, portfolio_value=1.07, market_value=1.054, episode=1.138e+04\n", + "loss=6.037e-06, reward=6.095e-06, cash_bias=0.1787, portfolio_value=1.057, market_value=1.05, episode=1.139e+04\n", + "loss=6.038e-06, reward=-2.846e-06, cash_bias=0.2016, portfolio_value=1.016, market_value=1.011, episode=1.141e+04\n", + "loss=6.023e-06, reward=-2.276e-05, cash_bias=0.2031, portfolio_value=1.068, market_value=1.05, episode=1.142e+04\n", + "loss=6.055e-06, reward=-8.815e-06, cash_bias=0.1702, portfolio_value=1.02, market_value=1.028, episode=1.144e+04\n", + "loss=6.051e-06, reward=4.421e-06, cash_bias=0.04792, portfolio_value=1.054, market_value=1.049, episode=1.146e+04\n", + "loss=6.175e-06, reward=7.054e-06, cash_bias=0.1361, portfolio_value=1.072, market_value=1.056, episode=1.147e+04\n", + "loss=6.465e-06, reward=-2.024e-06, cash_bias=0.2457, portfolio_value=1.058, market_value=1.028, episode=1.149e+04\n", + "loss=6.64e-06, reward=-1.377e-05, cash_bias=0.1815, portfolio_value=1.049, market_value=1.041, episode=1.15e+04\n", + "loss=6.845e-06, reward=1.155e-06, cash_bias=0.2623, portfolio_value=1.057, market_value=1.018, episode=1.152e+04\n", + "loss=6.733e-06, reward=-2.934e-05, cash_bias=0.09789, portfolio_value=1.03, market_value=1.037, episode=1.154e+04\n", + "loss=6.732e-06, reward=6.837e-06, cash_bias=0.1413, portfolio_value=1.05, market_value=1.032, episode=1.155e+04\n", + "loss=6.426e-06, reward=7.446e-06, cash_bias=0.1318, portfolio_value=1.041, market_value=1.034, episode=1.157e+04\n", + "loss=6.604e-06, reward=-1.391e-05, cash_bias=0.04306, portfolio_value=1.05, market_value=1.051, episode=1.158e+04\n", + "loss=6.54e-06, reward=1.726e-05, cash_bias=0.1906, portfolio_value=1.09, market_value=1.042, episode=1.16e+04\n", + "loss=6.456e-06, reward=5.106e-06, cash_bias=0.0848, portfolio_value=1.032, market_value=1.013, episode=1.162e+04\n", + "loss=6.625e-06, reward=-5.3e-05, cash_bias=0.2593, portfolio_value=1.048, market_value=1.057, episode=1.163e+04\n", + "loss=6.34e-06, reward=6.536e-06, cash_bias=0.1747, portfolio_value=1.063, market_value=1.024, episode=1.165e+04\n", + "loss=6.368e-06, reward=-3.593e-06, cash_bias=0.1922, portfolio_value=1.023, market_value=1.029, episode=1.166e+04\n", + "loss=6.263e-06, reward=2.123e-05, cash_bias=0.1657, portfolio_value=1.067, market_value=1.041, episode=1.168e+04\n", + "loss=6.185e-06, reward=8.537e-06, cash_bias=0.163, portfolio_value=1.094, market_value=1.032, episode=1.17e+04\n", + "loss=6.196e-06, reward=2.967e-05, cash_bias=0.1897, portfolio_value=1.071, market_value=1.038, episode=1.171e+04\n", + "loss=6.039e-06, reward=6.356e-06, cash_bias=0.09591, portfolio_value=1.088, market_value=1.036, episode=1.173e+04\n", + "loss=5.923e-06, reward=-1.244e-06, cash_bias=0.1077, portfolio_value=1.07, market_value=1.012, episode=1.174e+04\n", + "loss=5.809e-06, reward=-1.56e-06, cash_bias=0.05822, portfolio_value=1.014, market_value=1.022, episode=1.176e+04\n", + "loss=5.98e-06, reward=-1.021e-05, cash_bias=0.1095, portfolio_value=1.045, market_value=1.032, episode=1.178e+04\n", + "loss=5.877e-06, reward=-6.796e-06, cash_bias=0.1365, portfolio_value=1.062, market_value=1.028, episode=1.179e+04\n", + "loss=5.883e-06, reward=2.397e-05, cash_bias=0.1529, portfolio_value=1.088, market_value=1.05, episode=1.181e+04\n", + "loss=5.903e-06, reward=6.54e-06, cash_bias=0.1102, portfolio_value=1.083, market_value=1.054, episode=1.182e+04\n", + "loss=5.908e-06, reward=6.154e-06, cash_bias=0.08573, portfolio_value=1.051, market_value=1.037, episode=1.184e+04\n", + "loss=5.707e-06, reward=-1.599e-05, cash_bias=0.05663, portfolio_value=1.015, market_value=1.031, episode=1.186e+04\n", + "loss=5.275e-06, reward=2.576e-05, cash_bias=0.1075, portfolio_value=1.032, market_value=1.031, episode=1.187e+04\n", + "loss=5.302e-06, reward=1.103e-05, cash_bias=0.1293, portfolio_value=1.064, market_value=1.024, episode=1.189e+04\n", + "loss=5.229e-06, reward=-2.173e-07, cash_bias=0.1786, portfolio_value=1.032, market_value=1.024, episode=1.19e+04\n", + "loss=5.089e-06, reward=-1.036e-05, cash_bias=0.2089, portfolio_value=1.048, market_value=1.032, episode=1.192e+04\n", + "loss=5.19e-06, reward=1.185e-05, cash_bias=0.1645, portfolio_value=1.054, market_value=1.043, episode=1.194e+04\n", + "loss=5.173e-06, reward=3.501e-06, cash_bias=0.1648, portfolio_value=1.038, market_value=1.031, episode=1.195e+04\n", + "loss=4.9e-06, reward=-4.886e-07, cash_bias=0.1358, portfolio_value=1.036, market_value=1.023, episode=1.197e+04\n", + "loss=4.554e-06, reward=1.935e-06, cash_bias=0.1028, portfolio_value=1.078, market_value=1.022, episode=1.198e+04\n", + "loss=4.383e-06, reward=-8.587e-09, cash_bias=0.1746, portfolio_value=1.048, market_value=1.019, episode=1.2e+04\n", + "loss=4.22e-06, reward=-4.032e-06, cash_bias=0.0744, portfolio_value=1.057, market_value=1.022, episode=1.202e+04\n", + "loss=4.217e-06, reward=2.723e-06, cash_bias=0.1202, portfolio_value=1.074, market_value=1.059, episode=1.203e+04\n", + "loss=4.114e-06, reward=1.845e-05, cash_bias=0.1968, portfolio_value=1.072, market_value=1.043, episode=1.205e+04\n", + "loss=3.851e-06, reward=1.082e-05, cash_bias=0.09763, portfolio_value=1.047, market_value=1.018, episode=1.206e+04\n", + "loss=3.9e-06, reward=1.52e-05, cash_bias=0.09043, portfolio_value=1.029, market_value=1.015, episode=1.208e+04\n", + "loss=3.723e-06, reward=1.632e-05, cash_bias=0.1069, portfolio_value=1.057, market_value=1.034, episode=1.21e+04\n", + "loss=3.552e-06, reward=-1.148e-05, cash_bias=0.1942, portfolio_value=1.03, market_value=1.014, episode=1.211e+04\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=3.491e-06, reward=1.384e-05, cash_bias=0.1383, portfolio_value=1.043, market_value=1.038, episode=1.213e+04\n", + "loss=3.495e-06, reward=-3.243e-06, cash_bias=0.1496, portfolio_value=1.006, market_value=1.008, episode=1.214e+04\n", + "loss=3.619e-06, reward=2.108e-05, cash_bias=0.1574, portfolio_value=1.082, market_value=1.043, episode=1.216e+04\n", + "loss=3.52e-06, reward=5.574e-06, cash_bias=0.147, portfolio_value=1.086, market_value=1.051, episode=1.218e+04\n", + "loss=3.516e-06, reward=-1.822e-05, cash_bias=0.1789, portfolio_value=1.083, market_value=1.043, episode=1.219e+04\n", + "loss=3.538e-06, reward=-2.353e-06, cash_bias=0.1785, portfolio_value=1.076, market_value=1.055, episode=1.221e+04\n", + "loss=3.55e-06, reward=-2.098e-05, cash_bias=0.09693, portfolio_value=1.035, market_value=1.027, episode=1.222e+04\n", + "loss=3.683e-06, reward=-5.486e-07, cash_bias=0.1077, portfolio_value=1.08, market_value=1.029, episode=1.224e+04\n", + "loss=3.567e-06, reward=1.943e-05, cash_bias=0.139, portfolio_value=1.096, market_value=1.042, episode=1.226e+04\n", + "loss=3.456e-06, reward=-5.177e-06, cash_bias=0.1464, portfolio_value=1.05, market_value=1.04, episode=1.227e+04\n", + "loss=3.486e-06, reward=3.832e-06, cash_bias=0.06106, portfolio_value=1.055, market_value=1.026, episode=1.229e+04\n", + "loss=3.542e-06, reward=-3.935e-06, cash_bias=0.09939, portfolio_value=1.054, market_value=1.03, episode=1.23e+04\n", + "loss=3.549e-06, reward=-1.457e-05, cash_bias=0.03977, portfolio_value=1.08, market_value=1.055, episode=1.232e+04\n", + "loss=3.682e-06, reward=3.036e-05, cash_bias=0.2795, portfolio_value=1.079, market_value=1.049, episode=1.234e+04\n", + "loss=3.8e-06, reward=-1.336e-05, cash_bias=0.1788, portfolio_value=1.046, market_value=1.017, episode=1.235e+04\n", + "loss=3.805e-06, reward=-1.815e-05, cash_bias=0.1485, portfolio_value=1.057, market_value=1.028, episode=1.237e+04\n", + "loss=3.634e-06, reward=1.105e-05, cash_bias=0.1383, portfolio_value=1.008, market_value=1.005, episode=1.238e+04\n", + "loss=3.85e-06, reward=-1.268e-05, cash_bias=0.1193, portfolio_value=1.052, market_value=1.024, episode=1.24e+04\n", + "loss=3.773e-06, reward=2.157e-06, cash_bias=0.1904, portfolio_value=1.012, market_value=1.012, episode=1.242e+04\n", + "loss=3.763e-06, reward=-1.317e-05, cash_bias=0.1638, portfolio_value=1.064, market_value=1.056, episode=1.243e+04\n", + "loss=3.714e-06, reward=-8.163e-06, cash_bias=0.09744, portfolio_value=1.056, market_value=1.026, episode=1.245e+04\n", + "loss=3.654e-06, reward=-1.459e-05, cash_bias=0.2102, portfolio_value=1.077, market_value=1.051, episode=1.246e+04\n", + "loss=3.652e-06, reward=-9.936e-06, cash_bias=0.128, portfolio_value=1.061, market_value=1.038, episode=1.248e+04\n", + "loss=3.54e-06, reward=1.509e-05, cash_bias=0.1664, portfolio_value=1.064, market_value=1.026, episode=1.25e+04\n", + "loss=3.553e-06, reward=5.736e-06, cash_bias=0.222, portfolio_value=1.043, market_value=1.019, episode=1.251e+04\n", + "loss=3.536e-06, reward=-9.033e-06, cash_bias=0.1244, portfolio_value=1.08, market_value=1.028, episode=1.253e+04\n", + "loss=3.467e-06, reward=1.309e-05, cash_bias=0.2076, portfolio_value=1.048, market_value=1.033, episode=1.254e+04\n", + "loss=3.472e-06, reward=4.669e-05, cash_bias=0.09313, portfolio_value=1.046, market_value=1.028, episode=1.256e+04\n", + "loss=3.609e-06, reward=1.19e-05, cash_bias=0.2019, portfolio_value=1.104, market_value=1.058, episode=1.258e+04\n", + "loss=3.514e-06, reward=-8.675e-06, cash_bias=0.2057, portfolio_value=1.088, market_value=1.038, episode=1.259e+04\n", + "loss=3.652e-06, reward=2.28e-06, cash_bias=0.06489, portfolio_value=1.092, market_value=1.05, episode=1.261e+04\n", + "loss=3.604e-06, reward=-3.785e-06, cash_bias=0.1107, portfolio_value=1.059, market_value=1.032, episode=1.262e+04\n", + "loss=3.641e-06, reward=1.664e-06, cash_bias=0.06975, portfolio_value=1.091, market_value=1.04, episode=1.264e+04\n", + "loss=3.563e-06, reward=1.434e-05, cash_bias=0.1671, portfolio_value=1.058, market_value=1.027, episode=1.266e+04\n", + "loss=3.656e-06, reward=-5.185e-06, cash_bias=0.1525, portfolio_value=0.9992, market_value=1.026, episode=1.267e+04\n", + "loss=3.75e-06, reward=7.216e-07, cash_bias=0.09263, portfolio_value=1.066, market_value=1.015, episode=1.269e+04\n", + "loss=3.889e-06, reward=-3.538e-06, cash_bias=0.2271, portfolio_value=1.101, market_value=1.045, episode=1.27e+04\n", + "loss=3.779e-06, reward=2.559e-05, cash_bias=0.1477, portfolio_value=1.041, market_value=1.048, episode=1.272e+04\n", + "loss=3.66e-06, reward=3.633e-05, cash_bias=0.09927, portfolio_value=1.044, market_value=1.025, episode=1.274e+04\n", + "loss=3.761e-06, reward=-1.445e-06, cash_bias=0.1012, portfolio_value=1.026, market_value=1.023, episode=1.275e+04\n", + "loss=3.659e-06, reward=-1.767e-05, cash_bias=0.1214, portfolio_value=1.043, market_value=1.054, episode=1.277e+04\n", + "loss=3.776e-06, reward=9.893e-06, cash_bias=0.1943, portfolio_value=1.071, market_value=1.061, episode=1.278e+04\n", + "loss=3.989e-06, reward=7.227e-06, cash_bias=0.1905, portfolio_value=1.052, market_value=1.048, episode=1.28e+04\n", + "loss=4.04e-06, reward=2.14e-05, cash_bias=0.1219, portfolio_value=1.065, market_value=1.033, episode=1.282e+04\n", + "loss=4.262e-06, reward=5.487e-06, cash_bias=0.1058, portfolio_value=1.088, market_value=1.054, episode=1.283e+04\n", + "loss=4.19e-06, reward=7.069e-06, cash_bias=0.1517, portfolio_value=1.013, market_value=1.021, episode=1.285e+04\n", + "loss=4.17e-06, reward=8.999e-06, cash_bias=0.2005, portfolio_value=1.06, market_value=1.024, episode=1.286e+04\n", + "loss=4.02e-06, reward=2.176e-05, cash_bias=0.1014, portfolio_value=1.037, market_value=1.008, episode=1.288e+04\n", + "loss=4.034e-06, reward=-1.025e-06, cash_bias=0.1832, portfolio_value=1.07, market_value=1.059, episode=1.29e+04\n", + "loss=4.223e-06, reward=-6.426e-06, cash_bias=0.1221, portfolio_value=1.082, market_value=1.045, episode=1.291e+04\n", + "loss=4.281e-06, reward=-1.411e-05, cash_bias=0.1563, portfolio_value=1.008, market_value=0.999, episode=1.293e+04\n", + "loss=4.227e-06, reward=-3.325e-07, cash_bias=0.1153, portfolio_value=1.059, market_value=1.023, episode=1.294e+04\n", + "loss=4.106e-06, reward=5.373e-06, cash_bias=0.04955, portfolio_value= 1, market_value=1.018, episode=1.296e+04\n", + "loss=3.869e-06, reward=-1.556e-05, cash_bias=0.126, portfolio_value=1.009, market_value=0.9984, episode=1.298e+04\n", + "loss=3.801e-06, reward=2.543e-05, cash_bias=0.1246, portfolio_value=1.067, market_value=1.056, episode=1.299e+04\n", + "loss=3.812e-06, reward=-1.092e-05, cash_bias=0.1343, portfolio_value=1.091, market_value=1.06, episode=1.301e+04\n", + "loss=3.887e-06, reward=7.976e-06, cash_bias=0.05123, portfolio_value=1.033, market_value=1.026, episode=1.302e+04\n", + "loss=3.993e-06, reward=-3.007e-05, cash_bias=0.1395, portfolio_value=1.048, market_value=1.017, episode=1.304e+04\n", + "loss=3.892e-06, reward=-1.123e-05, cash_bias=0.1462, portfolio_value=1.018, market_value=1.036, episode=1.306e+04\n", + "loss=3.788e-06, reward=1.695e-05, cash_bias=0.1145, portfolio_value=1.041, market_value=1.049, episode=1.307e+04\n", + "loss=3.765e-06, reward=-3.02e-06, cash_bias=0.2677, portfolio_value=1.075, market_value=1.038, episode=1.309e+04\n", + "loss=3.847e-06, reward=3.387e-06, cash_bias=0.2257, portfolio_value=1.046, market_value=1.008, episode=1.31e+04\n", + "loss=3.686e-06, reward=1.033e-06, cash_bias=0.07374, portfolio_value=1.023, market_value=1.012, episode=1.312e+04\n", + "loss=3.594e-06, reward=1.838e-05, cash_bias=0.2317, portfolio_value=1.1, market_value=1.061, episode=1.314e+04\n", + "loss=3.446e-06, reward=-2.664e-05, cash_bias=0.1217, portfolio_value=1.07, market_value=1.037, episode=1.315e+04\n", + "loss=3.39e-06, reward=-1.617e-06, cash_bias=0.1328, portfolio_value=1.098, market_value=1.061, episode=1.317e+04\n", + "loss=3.326e-06, reward=-6.036e-06, cash_bias=0.07084, portfolio_value=1.063, market_value=1.056, episode=1.318e+04\n", + "loss=3.27e-06, reward=-5.697e-06, cash_bias=0.159, portfolio_value=1.065, market_value=1.036, episode=1.32e+04\n", + "loss=3.313e-06, reward=-5.662e-06, cash_bias=0.1097, portfolio_value=1.026, market_value=1.014, episode=1.322e+04\n", + "loss=3.335e-06, reward=5.543e-06, cash_bias=0.1615, portfolio_value=1.077, market_value=1.006, episode=1.323e+04\n", + "loss=3.547e-06, reward=2.088e-06, cash_bias=0.1518, portfolio_value=1.049, market_value=1.028, episode=1.325e+04\n", + "loss=3.6e-06, reward=-1.251e-05, cash_bias=0.07125, portfolio_value=1.043, market_value=1.042, episode=1.326e+04\n", + "loss=3.549e-06, reward=1.175e-05, cash_bias=0.1975, portfolio_value=1.006, market_value=0.9987, episode=1.328e+04\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=3.642e-06, reward=-1.105e-05, cash_bias=0.09517, portfolio_value=1.001, market_value=1.007, episode=1.33e+04\n", + "loss=3.597e-06, reward=2.368e-05, cash_bias=0.1036, portfolio_value=1.073, market_value=1.033, episode=1.331e+04\n", + "loss=3.507e-06, reward=1.222e-05, cash_bias=0.1147, portfolio_value=1.055, market_value=1.042, episode=1.333e+04\n", + "loss=3.268e-06, reward=-1.382e-05, cash_bias=0.2563, portfolio_value=1.058, market_value=1.038, episode=1.334e+04\n", + "loss=3.126e-06, reward=-3.025e-05, cash_bias=0.129, portfolio_value=1.101, market_value=1.057, episode=1.336e+04\n", + "loss=3.082e-06, reward=-1.397e-06, cash_bias=0.1335, portfolio_value=1.031, market_value=1.032, episode=1.338e+04\n", + "loss=3.155e-06, reward=8.994e-06, cash_bias=0.1335, portfolio_value=1.028, market_value=1.011, episode=1.339e+04\n", + "loss=3.312e-06, reward=1.259e-05, cash_bias=0.06334, portfolio_value=1.033, market_value=1.033, episode=1.341e+04\n", + "loss=3.444e-06, reward=1.008e-05, cash_bias=0.09913, portfolio_value=1.045, market_value=1.043, episode=1.342e+04\n", + "loss=3.62e-06, reward=9.74e-06, cash_bias=0.1948, portfolio_value=1.075, market_value=1.037, episode=1.344e+04\n", + "loss=3.574e-06, reward=7.013e-06, cash_bias=0.06574, portfolio_value=1.06, market_value=1.042, episode=1.346e+04\n", + "loss=3.559e-06, reward=1.543e-05, cash_bias=0.1212, portfolio_value=1.053, market_value=1.037, episode=1.347e+04\n", + "loss=3.477e-06, reward=-2.222e-05, cash_bias=0.08112, portfolio_value=1.046, market_value=1.027, episode=1.349e+04\n", + "loss=3.448e-06, reward=-1.605e-05, cash_bias=0.07382, portfolio_value=1.058, market_value=1.053, episode=1.35e+04\n", + "loss=3.308e-06, reward=-1.801e-06, cash_bias=0.1376, portfolio_value=1.059, market_value=1.037, episode=1.352e+04\n", + "loss=3.107e-06, reward=-2.021e-05, cash_bias=0.1247, portfolio_value=1.05, market_value=1.027, episode=1.354e+04\n", + "loss=2.992e-06, reward=1.647e-05, cash_bias=0.1286, portfolio_value=1.03, market_value=1.011, episode=1.355e+04\n", + "loss=2.793e-06, reward=1.059e-05, cash_bias=0.1732, portfolio_value=1.049, market_value=1.013, episode=1.357e+04\n", + "loss=2.868e-06, reward=-9.779e-06, cash_bias=0.2205, portfolio_value=1.032, market_value=1.014, episode=1.358e+04\n", + "loss=2.957e-06, reward=5.736e-06, cash_bias=0.1086, portfolio_value=1.062, market_value=1.046, episode=1.36e+04\n", + "loss=3.087e-06, reward=9.475e-06, cash_bias=0.1077, portfolio_value=1.083, market_value=1.026, episode=1.362e+04\n", + "loss=2.986e-06, reward=8.419e-06, cash_bias=0.178, portfolio_value=1.037, market_value=1.015, episode=1.363e+04\n", + "loss=2.885e-06, reward=2.412e-05, cash_bias=0.1493, portfolio_value=1.033, market_value=1.018, episode=1.365e+04\n", + "loss=2.863e-06, reward=-1.151e-05, cash_bias=0.08362, portfolio_value=1.011, market_value=1.027, episode=1.366e+04\n", + "loss=2.882e-06, reward=4.348e-05, cash_bias=0.1481, portfolio_value=1.103, market_value=1.046, episode=1.368e+04\n", + "loss=2.984e-06, reward=5.769e-05, cash_bias=0.1954, portfolio_value=1.072, market_value=1.066, episode=1.37e+04\n", + "loss=2.899e-06, reward=8.188e-06, cash_bias=0.08994, portfolio_value=1.055, market_value=1.025, episode=1.371e+04\n", + "loss=2.984e-06, reward=8.083e-06, cash_bias=0.09102, portfolio_value=1.092, market_value=1.072, episode=1.373e+04\n", + "loss=2.899e-06, reward=2.004e-05, cash_bias=0.165, portfolio_value=1.035, market_value=1.035, episode=1.374e+04\n", + "loss=2.86e-06, reward=2.309e-05, cash_bias=0.07076, portfolio_value=1.053, market_value=1.031, episode=1.376e+04\n", + "loss=2.969e-06, reward=1.177e-05, cash_bias=0.06892, portfolio_value=1.054, market_value=1.035, episode=1.378e+04\n", + "loss=2.901e-06, reward=-1.481e-05, cash_bias=0.1034, portfolio_value=1.051, market_value=1.023, episode=1.379e+04\n", + "loss=2.839e-06, reward=-3.043e-05, cash_bias=0.1106, portfolio_value=1.064, market_value=1.035, episode=1.381e+04\n", + "loss=2.762e-06, reward=-6.886e-06, cash_bias=0.2255, portfolio_value=1.05, market_value=1.032, episode=1.382e+04\n", + "loss=2.755e-06, reward=2.912e-06, cash_bias=0.1205, portfolio_value=1.061, market_value=1.038, episode=1.384e+04\n", + "loss=2.687e-06, reward=-6.381e-06, cash_bias=0.06027, portfolio_value=1.021, market_value=1.025, episode=1.386e+04\n", + "loss=2.872e-06, reward=-6.835e-08, cash_bias=0.1521, portfolio_value=1.07, market_value=1.051, episode=1.387e+04\n", + "loss=2.753e-06, reward=1.132e-06, cash_bias=0.1518, portfolio_value=1.03, market_value=1.011, episode=1.389e+04\n", + "loss=2.82e-06, reward=-7.482e-07, cash_bias=0.0889, portfolio_value=1.062, market_value=1.03, episode=1.39e+04\n", + "loss=2.692e-06, reward=1.317e-05, cash_bias=0.1523, portfolio_value=1.066, market_value=1.039, episode=1.392e+04\n", + "loss=2.678e-06, reward=-2.915e-06, cash_bias=0.1473, portfolio_value=1.063, market_value=1.052, episode=1.394e+04\n", + "loss=2.744e-06, reward=6.31e-06, cash_bias=0.1444, portfolio_value=1.057, market_value=1.047, episode=1.395e+04\n", + "loss=2.635e-06, reward=-3.338e-05, cash_bias=0.1363, portfolio_value=1.014, market_value=0.9913, episode=1.397e+04\n", + "loss=2.622e-06, reward=-1.485e-05, cash_bias=0.04997, portfolio_value=1.044, market_value=1.023, episode=1.398e+04\n", + "loss=2.748e-06, reward=4.415e-06, cash_bias=0.06646, portfolio_value=1.072, market_value=1.046, episode=1.4e+04\n", + "loss=2.697e-06, reward=-2.424e-05, cash_bias=0.08241, portfolio_value=1.058, market_value=1.054, episode=1.402e+04\n", + "loss=2.721e-06, reward=2.583e-05, cash_bias=0.09391, portfolio_value=1.105, market_value=1.041, episode=1.403e+04\n", + "loss=2.57e-06, reward=2.145e-05, cash_bias=0.1663, portfolio_value=1.029, market_value=1.005, episode=1.405e+04\n", + "loss=2.594e-06, reward=-1.766e-05, cash_bias=0.06388, portfolio_value=1.045, market_value=1.021, episode=1.406e+04\n", + "loss=2.612e-06, reward=-9.027e-06, cash_bias=0.1639, portfolio_value=1.048, market_value=1.027, episode=1.408e+04\n", + "loss=2.664e-06, reward=-3.298e-05, cash_bias=0.1539, portfolio_value=1.026, market_value=1.015, episode=1.41e+04\n", + "loss=2.696e-06, reward=8.015e-06, cash_bias=0.1153, portfolio_value=1.041, market_value=1.019, episode=1.411e+04\n", + "loss=2.757e-06, reward=1.649e-05, cash_bias=0.09608, portfolio_value=1.062, market_value=1.027, episode=1.413e+04\n", + "loss=2.867e-06, reward=-1.215e-05, cash_bias=0.05146, portfolio_value=1.048, market_value=1.034, episode=1.414e+04\n", + "loss=2.879e-06, reward=-2.683e-06, cash_bias=0.1769, portfolio_value=1.072, market_value=1.033, episode=1.416e+04\n", + "loss=3.074e-06, reward=9.872e-06, cash_bias=0.1459, portfolio_value=1.057, market_value=1.036, episode=1.418e+04\n", + "loss=3.148e-06, reward=1.066e-05, cash_bias=0.1972, portfolio_value=1.055, market_value=1.045, episode=1.419e+04\n", + "loss=3.082e-06, reward=2.502e-05, cash_bias=0.1148, portfolio_value=1.028, market_value=1.039, episode=1.421e+04\n", + "loss=3.014e-06, reward=-4.333e-06, cash_bias=0.1477, portfolio_value=1.002, market_value=0.9933, episode=1.422e+04\n", + "loss=3.153e-06, reward=7.692e-06, cash_bias=0.1235, portfolio_value=1.094, market_value=1.063, episode=1.424e+04\n", + "loss=3.078e-06, reward=-1.276e-05, cash_bias=0.1848, portfolio_value=1.048, market_value=1.024, episode=1.426e+04\n", + "loss=3.28e-06, reward=9.311e-06, cash_bias=0.1037, portfolio_value=1.045, market_value=1.018, episode=1.427e+04\n", + "loss=3.485e-06, reward=1.2e-05, cash_bias=0.1167, portfolio_value=1.024, market_value=1.022, episode=1.429e+04\n", + "loss=3.649e-06, reward=-5.43e-06, cash_bias=0.1448, portfolio_value=1.05, market_value=1.06, episode=1.43e+04\n", + "loss=3.756e-06, reward=-2.282e-06, cash_bias=0.1114, portfolio_value=1.086, market_value=1.043, episode=1.432e+04\n", + "loss=3.882e-06, reward=1.782e-05, cash_bias=0.1457, portfolio_value=1.046, market_value=1.016, episode=1.434e+04\n", + "loss=3.749e-06, reward=6.492e-07, cash_bias=0.1834, portfolio_value=1.047, market_value=1.019, episode=1.435e+04\n", + "loss=3.555e-06, reward=-1.521e-05, cash_bias=0.1305, portfolio_value=1.054, market_value=1.034, episode=1.437e+04\n", + "loss=3.6e-06, reward=-5.092e-06, cash_bias=0.193, portfolio_value=1.022, market_value=1.037, episode=1.438e+04\n", + "loss=3.396e-06, reward=1.212e-05, cash_bias=0.1599, portfolio_value=1.052, market_value=1.036, episode=1.44e+04\n", + "loss=3.529e-06, reward=1.266e-07, cash_bias=0.05838, portfolio_value=1.047, market_value=1.022, episode=1.442e+04\n", + "loss=3.522e-06, reward=1.243e-05, cash_bias=0.1853, portfolio_value=1.052, market_value=1.014, episode=1.443e+04\n", + "loss=3.51e-06, reward=8.362e-06, cash_bias=0.03335, portfolio_value=1.018, market_value=1.022, episode=1.445e+04\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=3.623e-06, reward=7.438e-06, cash_bias=0.1662, portfolio_value=1.041, market_value=1.027, episode=1.446e+04\n", + "loss=3.938e-06, reward=-1.872e-05, cash_bias=0.1177, portfolio_value=1.105, market_value=1.064, episode=1.448e+04\n", + "loss=4.048e-06, reward=-5.486e-06, cash_bias=0.2364, portfolio_value=1.036, market_value=1.027, episode=1.45e+04\n", + "loss=4.344e-06, reward=-1.308e-05, cash_bias=0.1578, portfolio_value=1.054, market_value=1.038, episode=1.451e+04\n", + "loss=4.62e-06, reward=2.137e-05, cash_bias=0.2239, portfolio_value=1.05, market_value=1.041, episode=1.453e+04\n", + "loss=4.543e-06, reward=-8.339e-06, cash_bias=0.1255, portfolio_value=1.071, market_value=1.054, episode=1.454e+04\n", + "loss=4.261e-06, reward=1.463e-06, cash_bias=0.1752, portfolio_value=1.055, market_value=1.029, episode=1.456e+04\n", + "loss=4.321e-06, reward=4.942e-06, cash_bias=0.2023, portfolio_value=1.093, market_value=1.04, episode=1.458e+04\n", + "loss=4.21e-06, reward=-5.981e-06, cash_bias=0.1144, portfolio_value=1.042, market_value=1.041, episode=1.459e+04\n", + "loss=4.029e-06, reward=-4.522e-06, cash_bias=0.1183, portfolio_value=1.05, market_value=1.017, episode=1.461e+04\n", + "loss=4.166e-06, reward=-5.963e-06, cash_bias=0.147, portfolio_value=1.058, market_value=1.026, episode=1.462e+04\n", + "loss=4.232e-06, reward=-6.416e-07, cash_bias=0.1563, portfolio_value=1.048, market_value=1.032, episode=1.464e+04\n", + "loss=4.11e-06, reward=5.929e-06, cash_bias=0.2567, portfolio_value=1.038, market_value=1.054, episode=1.466e+04\n", + "loss=4.081e-06, reward=1.128e-05, cash_bias=0.09795, portfolio_value=1.074, market_value=1.044, episode=1.467e+04\n", + "loss=4.076e-06, reward=1.596e-05, cash_bias=0.1315, portfolio_value=1.065, market_value=1.047, episode=1.469e+04\n", + "loss=4.156e-06, reward=1.257e-05, cash_bias=0.1869, portfolio_value=1.065, market_value=1.03, episode=1.47e+04\n", + "loss=4.337e-06, reward=-2.827e-07, cash_bias=0.06197, portfolio_value=1.086, market_value=1.063, episode=1.472e+04\n", + "loss=4.233e-06, reward=5.22e-06, cash_bias=0.1377, portfolio_value=1.046, market_value=1.018, episode=1.474e+04\n", + "loss=4.199e-06, reward=1.6e-05, cash_bias=0.1644, portfolio_value=1.081, market_value=1.037, episode=1.475e+04\n", + "loss=3.903e-06, reward=4.384e-06, cash_bias=0.06146, portfolio_value=1.049, market_value=1.021, episode=1.477e+04\n", + "loss=3.828e-06, reward=9.488e-06, cash_bias=0.03033, portfolio_value=1.026, market_value=1.009, episode=1.478e+04\n", + "loss=4.106e-06, reward=1.292e-05, cash_bias=0.1179, portfolio_value=1.072, market_value=1.014, episode=1.48e+04\n", + "loss=4.291e-06, reward=1.26e-05, cash_bias=0.2138, portfolio_value=1.012, market_value=1.039, episode=1.482e+04\n", + "loss=4.124e-06, reward=-2.081e-06, cash_bias=0.0719, portfolio_value=1.036, market_value=1.03, episode=1.483e+04\n", + "loss=3.916e-06, reward=1.346e-05, cash_bias=0.1855, portfolio_value=1.054, market_value=1.036, episode=1.485e+04\n", + "loss=4.052e-06, reward=-9.327e-06, cash_bias=0.1553, portfolio_value=1.099, market_value=1.034, episode=1.486e+04\n", + "loss=3.934e-06, reward=4.464e-06, cash_bias=0.1326, portfolio_value=1.03, market_value=1.033, episode=1.488e+04\n", + "loss=3.812e-06, reward=6.633e-08, cash_bias=0.1452, portfolio_value=1.078, market_value=1.049, episode=1.49e+04\n", + "loss=3.309e-06, reward=-1.645e-06, cash_bias=0.1246, portfolio_value=1.046, market_value=1.022, episode=1.491e+04\n", + "loss=3.059e-06, reward=1.279e-05, cash_bias=0.1731, portfolio_value=1.01, market_value=0.9967, episode=1.493e+04\n", + "loss=2.561e-06, reward=1.703e-05, cash_bias=0.2322, portfolio_value=1.103, market_value=1.047, episode=1.494e+04\n", + "loss=1.516e-06, reward=-4.46e-06, cash_bias=0.1772, portfolio_value=1.054, market_value=1.031, episode=1.496e+04\n", + "loss=6.166e-07, reward=2.701e-07, cash_bias=0.1069, portfolio_value=1.088, market_value=1.069, episode=1.498e+04\n", + "loss=1.365e-07, reward=3.611e-05, cash_bias=0.1532, portfolio_value=1.064, market_value=1.063, episode=1.499e+04\n", + "loss=2.934e-08, reward=7.864e-08, cash_bias=0.1149, portfolio_value=1.057, market_value=1.017, episode=1.501e+04\n", + "loss=1.872e-08, reward=2.698e-05, cash_bias=0.1095, portfolio_value=1.029, market_value=1.026, episode=1.502e+04\n", + "loss=1.851e-08, reward=-5.23e-06, cash_bias=0.106, portfolio_value=1.031, market_value=1.027, episode=1.504e+04\n", + "loss=1.286e-08, reward=1.256e-05, cash_bias=0.1046, portfolio_value=1.051, market_value=1.022, episode=1.506e+04\n", + "loss=1.917e-08, reward=1.839e-05, cash_bias=0.1441, portfolio_value=1.084, market_value=1.035, episode=1.507e+04\n", + "loss=1.883e-08, reward=-5.764e-06, cash_bias=0.1603, portfolio_value=1.013, market_value=0.9794, episode=1.509e+04\n", + "loss=1.426e-08, reward=-1.582e-05, cash_bias=0.04607, portfolio_value=1.105, market_value=1.052, episode=1.51e+04\n", + "loss=2.495e-08, reward=2.211e-05, cash_bias=0.11, portfolio_value=1.024, market_value=1.023, episode=1.512e+04\n", + "loss=9.473e-09, reward=5.633e-06, cash_bias=0.06172, portfolio_value=1.044, market_value=1.033, episode=1.514e+04\n", + "loss=1.75e-08, reward=-1.363e-07, cash_bias=0.1696, portfolio_value=1.034, market_value=1.036, episode=1.515e+04\n", + "loss=2.587e-08, reward=-1.373e-05, cash_bias=0.1087, portfolio_value=1.073, market_value=1.046, episode=1.517e+04\n", + "loss=2.274e-08, reward=-1.419e-05, cash_bias=0.1931, portfolio_value=1.053, market_value=1.02, episode=1.518e+04\n", + "loss=2.611e-08, reward=8.005e-06, cash_bias=0.105, portfolio_value=1.072, market_value=1.032, episode=1.52e+04\n", + "loss=2.035e-08, reward=7.17e-06, cash_bias=0.1642, portfolio_value=1.054, market_value=1.026, episode=1.522e+04\n", + "loss=1.013e-08, reward=-4.191e-06, cash_bias=0.1328, portfolio_value=1.056, market_value=1.038, episode=1.523e+04\n", + "loss=2.776e-08, reward=-6.256e-06, cash_bias=0.04562, portfolio_value=1.066, market_value=1.036, episode=1.525e+04\n", + "loss=2.105e-08, reward=1.13e-05, cash_bias=0.1085, portfolio_value=1.08, market_value=1.059, episode=1.526e+04\n", + "loss=1.138e-08, reward=1.753e-05, cash_bias=0.1395, portfolio_value=1.034, market_value=1.03, episode=1.528e+04\n", + "loss=1.873e-08, reward=8.798e-06, cash_bias=0.1284, portfolio_value=1.051, market_value=1.027, episode=1.53e+04\n", + "loss=1.496e-08, reward=-4.759e-06, cash_bias=0.1092, portfolio_value=1.073, market_value=1.058, episode=1.531e+04\n", + "loss=1.475e-08, reward=1.567e-05, cash_bias=0.1789, portfolio_value=1.089, market_value=1.072, episode=1.533e+04\n", + "loss=2.23e-08, reward=-7.478e-06, cash_bias=0.1131, portfolio_value=1.087, market_value=1.031, episode=1.534e+04\n", + "loss=2.773e-08, reward=2.296e-06, cash_bias=0.2103, portfolio_value=1.095, market_value=1.071, episode=1.536e+04\n", + "loss=2.308e-08, reward=9.93e-06, cash_bias=0.07148, portfolio_value=1.066, market_value=1.029, episode=1.538e+04\n", + "loss=1.31e-08, reward=-2.737e-05, cash_bias=0.1271, portfolio_value=1.053, market_value=1.037, episode=1.539e+04\n", + "loss=1.785e-08, reward=-6.244e-06, cash_bias=0.06846, portfolio_value=1.047, market_value=1.028, episode=1.541e+04\n", + "loss=1.281e-08, reward=2.64e-05, cash_bias=0.1507, portfolio_value=1.07, market_value=1.051, episode=1.542e+04\n", + "loss=1.047e-08, reward=1.28e-05, cash_bias=0.1822, portfolio_value=1.031, market_value=1.037, episode=1.544e+04\n", + "loss=1.516e-08, reward=-7.513e-06, cash_bias=0.202, portfolio_value=1.032, market_value=1.016, episode=1.546e+04\n", + "loss=2.053e-08, reward=2.981e-05, cash_bias=0.1949, portfolio_value=1.088, market_value=1.036, episode=1.547e+04\n", + "loss=2.426e-08, reward=1.306e-05, cash_bias=0.06682, portfolio_value=1.045, market_value=1.033, episode=1.549e+04\n", + "loss=1.644e-08, reward=-2.807e-05, cash_bias=0.1116, portfolio_value=1.068, market_value=1.05, episode=1.55e+04\n", + "loss=2.031e-08, reward=-9.505e-06, cash_bias=0.06259, portfolio_value=1.001, market_value=1.023, episode=1.552e+04\n", + "loss=1.894e-08, reward=2.7e-05, cash_bias=0.1437, portfolio_value=1.009, market_value=1.006, episode=1.554e+04\n", + "loss=3.842e-08, reward=4.022e-06, cash_bias=0.1376, portfolio_value=1.068, market_value=1.02, episode=1.555e+04\n", + "loss=3.111e-08, reward=1.343e-05, cash_bias=0.1804, portfolio_value=1.062, market_value=1.026, episode=1.557e+04\n", + "loss=1.673e-08, reward=-3.933e-07, cash_bias=0.1303, portfolio_value=1.044, market_value=1.044, episode=1.558e+04\n", + "loss=1.597e-08, reward=3.65e-05, cash_bias=0.06151, portfolio_value=1.065, market_value=1.05, episode=1.56e+04\n", + "loss=1.907e-08, reward=4.902e-06, cash_bias=0.1022, portfolio_value=1.076, market_value=1.057, episode=1.562e+04\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "loss=4.412e-08, reward=3.156e-05, cash_bias=0.09332, portfolio_value=1.082, market_value=1.053, episode=1.563e+04\n", + "\n" + ] + } + ], + "source": [ + "memory = ReplayMemory(params.num_steps)\n", + "# memory = PrioritisedReplayMemory(params.num_steps)\n", + "\n", + "num_inputs = int(np.prod(env.observation_space.shape))\n", + "num_outputs = int(np.prod(env.action_space.shape))\n", + "\n", + "state = env.reset()\n", + "state = Variable(torch.Tensor(state).unsqueeze(0))\n", + "done = True\n", + "episode_length = 0\n", + "reports = []\n", + "\n", + "with tqdm(total=params.time_horizon, mininterval=2, unit='steps') as p:\n", + " episode = -1 \n", + " steps = 0\n", + " # horizon loop\n", + " while steps < params.time_horizon:\n", + " infos = []\n", + " episode_length = 0\n", + " # Sample data from the policy\n", + " while (len(memory.memory) < params.num_steps):\n", + " states = []\n", + " actions = []\n", + " rewards = []\n", + " values = []\n", + " returns = []\n", + " advantages = []\n", + " av_reward = 0\n", + " cum_reward = 0\n", + " cum_done = 0\n", + " # n steps loops\n", + " for step in range(params.num_steps):\n", + " # shared_obs_stats.observes(state)\n", + " # state = shared_obs_stats.normalize(state)\n", + " states.append(state)\n", + " \n", + " mu, sigma_sq, v = model(state)\n", + " eps = torch.randn(mu.size())\n", + " action = (mu + sigma_sq.sqrt() * Variable(eps))\n", + " env_action = action.data.squeeze().numpy()\n", + " state, reward, done, info = env.step(env_action)\n", + " done = (done or episode_length >= params.max_episode_length)\n", + " \n", + " cum_reward += reward\n", + " reward = max(min(reward, 1), -1)\n", + " rewards.append(reward)\n", + " actions.append(action)\n", + " values.append(v)\n", + " \n", + " steps+=1 \n", + " p.update(1)\n", + " if done:\n", + " episode += 1\n", + " cum_done += 1\n", + " av_reward += cum_reward\n", + " p.desc='av_reward={: 2.8f}'.format(av_reward / float(cum_done))\n", + " cum_reward = 0\n", + " episode_length = 0\n", + " infos.append(info)\n", + " state = env.reset()\n", + " \n", + " state = Variable(torch.Tensor(state).unsqueeze(0))\n", + " \n", + " if done:\n", + " break\n", + " \n", + " # one last step\n", + " R = torch.zeros(1, 1)\n", + " if not done:\n", + " _, _, v = model(state)\n", + " R = v.data\n", + " \n", + " # compute returns and GAE(lambda) advantages:\n", + " values.append(Variable(R))\n", + " R = Variable(R)\n", + " A = Variable(torch.zeros(1, 1))\n", + " for i in reversed(range(len(rewards))):\n", + " td = rewards[i] + params.gamma*values[i+1].data[0,0] - values[i].data[0,0]\n", + " A = float(td) + params.gamma * params.gae_param * A\n", + " advantages.insert(0, A)\n", + " R = A + values[i]\n", + " returns.insert(0, R)\n", + " \n", + " # store useful info:\n", + " memory.push([states, actions, returns, advantages])\n", + " \n", + "\n", + " # perform several epochs of optimization on the sampled data\n", + " model_old = Model(env.observation_space.shape,\n", + " env.action_space.shape)\n", + " model_old.load_state_dict(model.state_dict())\n", + " av_loss = 0\n", + " for k in range(params.num_epoch):\n", + " # cf https://github.com/openai/baselines/blob/master/baselines/pposgd/pposgd_simple.py\n", + " batch_states, batch_actions, batch_returns, batch_advantages = memory.sample(\n", + " params.batch_size)\n", + " \n", + " # old probas\n", + " mu_old, sigma_sq_old, v_pred_old = model_old(batch_states.detach())\n", + " probs_old = normal(batch_actions, mu_old, sigma_sq_old)\n", + " \n", + " # new probas\n", + " mu, sigma_sq, v_pred = model(batch_states)\n", + " probs = normal(batch_actions, mu, sigma_sq)\n", + " \n", + " # ratio\n", + " ratio = probs / (1e-15 + probs_old)\n", + " \n", + " # surrogate clip loss\n", + " surr1 = ratio * torch.cat([batch_advantages]*num_outputs,1) # surrogate from conservative policy iteration\n", + " surr2 = ratio.clamp(1-params.clip, 1+params.clip) * torch.cat([batch_advantages]*num_outputs,1)\n", + " loss_clip = -torch.mean(torch.min(surr1, surr2))\n", + " # should this be a mean along axis 0?\n", + " \n", + " # state-value function loss, do we even need this if they don't share params?\n", + " vfloss1 = (v_pred - batch_returns)**2\n", + " v_pred_clipped = v_pred_old + (v_pred - v_pred_old).clamp(-params.clip, params.clip)\n", + " vfloss2 = (v_pred_clipped - batch_returns)**2\n", + " loss_value = 0.5 * torch.mean(torch.max(vfloss1, vfloss2))\n", + " # should this be a mean along axis 0?\n", + " \n", + " # loss on entropy bonus to ensure sufficient exploration\n", + " loss_ent = -params.ent_coeff*torch.mean(probs*torch.log(probs+1e-5))\n", + " \n", + " # total\n", + " total_loss = (loss_clip + loss_value + loss_ent)\n", + "# total_loss = (loss_clip - loss_value + loss_ent)\n", + " av_loss += loss_value.data[0] / float(params.num_epoch)\n", + " \n", + " # before step, update old_model:\n", + " model_old.load_state_dict(model.state_dict())\n", + " \n", + " # step\n", + " optimizer.zero_grad()\n", + " total_loss.backward(retain_variables=True)\n", + " optimizer.step()\n", + " \n", + " # t finish, print:\n", + " df_infos = pd.DataFrame(infos)\n", + " \n", + " # show stats?\n", + "# display(df_infos[[\"cash_bias\",\"return\",\"portfolio_value\",\"market_value\"]].describe().loc[[\"min\",\"mean\",\"max\"]])\n", + " \n", + " report=OrderedDict(\n", + " episode=episode,\n", + "# reward=av_reward / float(cum_done),\n", + " loss=av_loss,\n", + " cash_bias=df_infos.cash_bias.mean(),\n", + " market_value=df_infos.market_value.mean(),\n", + " portfolio_value=df_infos.portfolio_value.mean(),\n", + " reward=df_infos.reward.mean()\n", + " )\n", + " \n", + " s = ', '.join(['{}={:2.4g}'.format(key,value) for key,value in report.items()])\n", + " print(s)\n", + " \n", + " reports.append(report)\n", + " \n", + " memory.clear()\n", + " torch.save(model_old, save_path)\n" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "start_time": "2017-08-05T23:33:26.049Z" + } + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T04:13:48.859942Z", + "start_time": "2017-08-06T04:13:48.114604Z" + } + }, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtUAAALICAYAAACw+wwuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# show progress\n", + "df=pd.DataFrame(reports)\n", + "g = sns.jointplot(x=\"episode\", y=\"loss\", data=df, kind=\"reg\", size=10)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-06T04:14:08.197294Z", + "start_time": "2017-08-06T04:13:48.861585Z" + }, + "scrolled": false + }, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEwCAYAAAC+DabKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQETEUQqAiIijfgCwDhLmDpoRjwAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEwCAYAAABbv6HjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nb/YrVpqZ3jd2AAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAEwCAYAAAC35gawAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEwCAYAAAC+DabKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noPx/Lmaf8NDsO1YAAAAASUVORK5CYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEwCAYAAABbv6HjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nmIhCX0TERBT6IiIm8h1OgWHg+iTdKQAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAEgCAYAAABRggMCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhULhxvx/KfI+9Hn+b/kAAAAASUVORK5CYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEwCAYAAABbv6HjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzsnXdYFOf2x7/vLrBLh6VJsWJFY4vdxN5SNKZoiumXGJNf\nrteoSbyam3JNjNc0S2ISTdTUG68pxpjE3pIYCyI2kCKCiCCwsLAsZcuc3x/v7sDCLizLroDO53l8\nZGbeeeed3XfPnDnnvOcwIiJISEhISNwQyFp6ABISEhIS1w5J6EtISEjcQEhCX0JCQuIGQhL6EhIS\nEjcQktCXkJCQuIGQhL6EhITEDYQk9JtJVlYWGGP4448/WnooAIDHH38cEyZMaOlhSDhIa5s/rYVO\nnTrhjTfeaOlhXJdIQl+iVZOdnY1Zs2YhNDQUSqUSPXr0wM8//9zgOfPmzcPQoUPh4+MDDw8Pm20e\nfvhhxMbGwtvbGyEhIZg4cSL++usvqzYvvvgiQkJCEBkZiVWrVlkdy83NRXR0NDIzM5t3gzc48fHx\nGDNmTL39x48fx/PPP3/Nx5OWlobJkyfDx8cHoaGhmDNnDnQ6XaPn/frrr+jfvz8UCgU6deqE9957\nz+r4yZMnMWbMGEREREChUKBDhw547rnnoNFo6vW1fv169OvXD0qlEiqVClOnThWPVVZW4s4770SH\nDh2gVCoRERGB6dOnIzk52eF7tP2LkHApBoMBnp6eLT2MNkdubi6GDRuGsWPHYvv27WjXrh2ys7MR\nEBDQ4HkmkwkPPfQQcnJysHLlSptthg0bhqeffhrt27dHWVkZ3n//fUycOBGpqamIjo7G9u3bsWnT\nJvz666/QaDSYPn06xo8fjz59+gAA5syZgxdffBFdunRx+X27ipacd829dlhYmAtH4xjl5eUYP348\n+vbti8OHD6O4uBhPPvkkNBoNvv32W7vnJSQk4K677sLChQvx3//+F0ePHsWcOXPg4+ODOXPmAAAU\nCgUef/xxDBgwAMHBwUhNTcX//d//ITc3Fz/++KPY17/+9S+sW7cOK1aswIgRI2AwGHD69GnxOGMM\nEydOxMsvv4zIyEgUFhbitddew/jx43Hx4kUolcrGb5RaIaNHj6Ynn3ySlixZQmFhYRQYGEiLFy8m\nk8lEr7/+OoWHh1NoaCgtXrzY6ryvv/6ahgwZQgEBARQSEkK33347paamisc3b95Mnp6edPToUXHf\n559/Tkqlkk6dOuXQ2DZv3kyxsbGkUCho+PDh9NNPPxEA+v3334mIaP/+/QSAtm/fTiNHjiSFQkFr\n166l4uJimjVrFrVv356USiV1796d3nnnHRIEgYiIMjIyCAClpaWJ1+rYsSNFR0eL22lpaQSAzp8/\nT0REarWaZs6cST4+PhQeHk5LliyhRx99lMaPHy+eo9fr6aWXXqKoqCjy9PSkXr160ddffy0ef/nl\nl2nEiBHi9r59+wgALVmyRNy3ePFiGjZsmNX97dq1i2699Vby9vamXr160a+//urQ59cUHn30UfG6\nzrBx40aSy+UOtdVoNASAtm7dSkREK1asoPvvv188PmTIEPrf//5HRERffvkljRw5kkwmk0N9//77\n7zRixAjy8/MjPz8/6tu3L+3YsYOIiC5evEgAaPPmzXTHHXeQt7c3de7cmTZu3GjVx8qVK6lfv37k\n6+tLERERdP/999OVK1fE4/bmneUz2L17N8XFxZFCoaAhQ4bQyZMnrfpPSEigiRMnkq+vL4WGhtLd\nd99NWVlZDt2fs3P+1VdfJQBW/yz33bFjR1q6dKl4jbKyMpo9ezaFhoaSl5cX3XzzzbRz506Hxuco\nn3zyCSmVStJoNOK+7du3EwDKzMy0e96DDz5Iw4cPt9q3cOFC6tixY4PXW7lyJQUFBYnbGRkZJJPJ\nxLnhKElJSQSAkpKSHGrfaoV+QEAAvfjii5SamkqfffYZAaApU6bQCy+8QKmpqbRp0yYCYCVsNmzY\nQNu2baOMjAxKTEykqVOnUteuXam6ulpsEx8fT126dKHS0lJKTU0lPz8/+vDDDx0aV2JiIslkMlq0\naBGdP3+evv/+e+rUqZNNod+jRw/atm0bZWZmUk5ODuXl5dFbb71FJ06coMzMTPryyy/J19eXNmzY\nIPbfoUMH+vjjj4mITwClUkl+fn7ig+vjjz+2eghMnz6dYmNjae/evXT27FmaNWsW+fv7Wwn9hQsX\nkkqlov/973+UmppKb775JjHGaM+ePUREtHfvXvL09CStVktE/CEQFhZmNYmHDRsmPmAt99e3b1/6\n7bffKC0tjR5//HHy9/en4uJi8Zy4uDjy9fVt8F92drbdz9pkMlFAQAA9//zz9MADD1BYWBj16dOH\nli1bRgaDwaHvy1GhX1lZScuWLSM/Pz/Kzc0lIqKdO3dS586dqbCwkC5cuEBBQUGUkpJC+fn5FB0d\nbaVMNITBYKDg4GB6/vnnKS0tjdLS0uiHH36gQ4cOEVGN0O/cuTNt3ryZ0tPT6Z///CfJ5XKra6xc\nuZJ2795NmZmZdPjwYRo+fDiNGjVKPG5v3m3cuJEYYzRgwAA6cOAAnTp1iu644w6KioqiiooKIiI6\nd+4c+fr60iuvvEIpKSl0+vRpuu+++6hbt25UWVnZ6D06O+e1Wi099NBDNHz4cMrLy6O8vDxxTHWF\n/n333UcdO3akHTt2UHJyMs2dO5c8PT0pJSVFbNPcOffoo4/S2LFjrfbp9XqSyWT05Zdf2j2vQ4cO\n9Prrr1vt27NnDwGgnJwcm+dcunSJbrnlFrr77rvFfW+//TYpFAr66quvKC4ujiIjI+n222+nM2fO\n2L12WVkZPffccxQdHU06nc5uu9q0WqHfr18/q31xcXHUp08fq319+/alBQsW2O1HrVYTAPrjjz/E\nfTqdjuLi4mjGjBnUv39/mj59usPjmjVrlpVWTES0Zs0am0L/iy++aLS/uXPn0oQJE8Ttxx57jGbM\nmEFEROvWraNx48bRbbfdRh999BEREc2cOZMefvhhIiJKT08XNW4L1dXVFBUVJQp9nU5HXl5e9R5q\n06dPFyd3ZWUlKRQK+uWXX4iIaMSIEfT222+LD4KysjLy8PAQHxKW+/v+++/F/vLz8wmAlYaSlZVF\n6enpDf5rSHhb+lQoFDR//nxKTEykr776ioKDg+u94dmjMaH/4Ycfkq+vLzHGKCYmxuoNkIho6dKl\n1LVrV+rZsyetX7+eiIjuvfde+s9//kNHjx6loUOHUrdu3eiNN96we43i4mICQPv377d53CL03333\nXXGf0WgkPz8/UQGwRWJiIgGgy5cvE5H9ebdx40YCIH5/ljH5+vrSp59+SkR83tV+qyEiqqqqIm9v\nb/rxxx/tjsFCc+b83/72Nxo9enS9drWFvmWuW+aohQEDBtATTzwhbjd3zk2cOJEefPDBevtDQ0Np\nxYoVds/z9PSkTz75xGrf2bNnCQAdO3bMav/w4cNJqVQSAJo2bZr4kCMimjNnDnl6elJsbCxt376d\njh07RjNmzKDg4GC6evWqVT8vvvgi+fr6EgDq1asXpaen2x1fXVqtTb9fv35W2+3atUO7du3q7Sso\nKBC3k5KS8PrrryMpKQlFRUUgcy657OxsjBw5EgDg4+ODzZs3o3///oiIiMDevXsdHlNycjLGjx9v\nte+WW26x2XbIkCFW24IgYMWKFfj2229x+fJlVFVVwWAwoGPHjmKbsWPH4oUXXgARYd++fRg/fjw8\nPT2xb98+PP300zhw4ACWL18ujgUARowYIZ7v5eWFwYMHo7y8HACQkZEBvV6PUaNGWY1l9OjReOut\ntwAASqUSw4cPx759+zBq1CgcP34cW7ZswcaNG3Ho0CEQEeRyufj5Wejfv7/4d0REBORyOa5evSru\nq31fjTFnzhx89dVX4nZycrJoD77pppvw7rvvAgAGDBiAvLw8LF26FG+++abD/dtj1qxZmDRpEgoK\nCrBu3Trcd999+OOPP9ChQwcAwMsvv4yXX35ZbP/dd9/h0qVL+PLLL9G9e3esX78ew4YNw+DBgzFw\n4EDcdttt9a4RHByM+Ph4TJ48GePGjcPo0aNx9913o0ePHlbtan+ecrkc4eHhVp/ngQMH8NZbbyE5\nORkajQaCIADgczs6OlpsV3feWRg+fLjVmHr16oVz584B4E7TjIwM+Pn5WZ1TVVWF9PT0hj/EWjgz\n5x3BMtfrzuNRo0ZZOd+b2m9LsHnzZpSXlyMlJQVLlizBnDlz8PnnnwPgn5fBYMCqVatwxx13AAC+\n+OILxMTE4KuvvsL8+fPFfl544QXEx8cjNzcX77zzDu6++24cPnwY/v7+jY6h1Ubv1HUCMcZs7rNM\n/oqKCkyaNAmMMWzcuBHHjh3D8ePHwRiDXq+3Os8SHldaWorCwkK3jN/X19dq+91338Vbb72FuXPn\nYvfu3UhKSkJ8fLzV2MaNG4fCwkKcPn0a+/fvx7hx4zBu3DgcOHAAZ86cQUFBAcaNG+fysY4bNw57\n9+7F77//ji5duiAqKgrjxo3Dvn37sG/fPgwfPryeg8jLy6teP5bvAgB69+4NPz+/Bv9dunQJAPDv\nf/8bSUlJ4r+oqCiEhobC09MTvXv3trpG7969UVZWhpKSkmbfd2BgILp27YoRI0Zg06ZN8PHxwdq1\na222LS4uxvz587Fx40akp6ejqqoKU6ZMQVBQEO666y7s2bPH7nXWr1+PEydOYOLEiTh48CD69OmD\nTz75xKpN3c+z9ty+dOkSbr/9dnTq1AnffvstEhISsG3bNgCoN7frzjtHEAQBjzzyiNV3kJSUhLS0\nNMTHxzvcjzNz3pU0Zc7ZIjIyEvn5+Vb7DAYDiouLERkZ2aTzLA/suue1b98evXr1wj333IMNGzbg\niy++wPnz563a1p7zSqUSsbGxyM7OtuonNDQU3bp1w5gxY/DDDz8gKysLX3/9td0x1qbVavpNJSUl\nBYWFhXjzzTfRq1cvAMDhw4dFbd/C2bNnMX/+fHz66afYunUrHnjgARw5cgQKhaLRa8TFxeHw4cNW\n+/7880+Hxnfo0CFMmTIFTz75pLivrhbVvn17xMbGYs2aNaisrMTgwYPBGIPRaMSqVavQpUsXUZuJ\ni4sT73HixIkAuAA4fvy4eP9du3aFQqHAoUOHxKgTAKLgsTB27Fi8+uqr2LJli/gmM27cOCxduhRE\nhHvuucehe6zNr7/+CoPB0GCbqKgoAEB4eDjCw8PrHR86dKj4g7CQmpqKwMBABAcHN3lMjSEIAqqq\nqmwemzt3LmbPno3evXsjKSkJRqMRRCQqFXK5vMG++/Tpgz59+mD+/PmYM2cO1q1bh6efftqhcR0/\nfhyVlZVYuXIlvL29AQAnTpxo0r0dOXJEVBg0Gg1SUlLE6w8aNAinT59GbGwsGGNN6rchHJnzXl5e\nMJlMDfZjEYKHDh3C7bffbtX/gAEDxO2mzDlbjBw5Ev/4xz9QVlYmRojt3r0bgiDUe9Ote97OnTvx\nyiuviPt27NiBjh07IiYmxu55loe6Zc7deuutAIDz58+jU6dOAPhv+uLFi3jggQcavC8isjt3bTVu\ndYwePZr+9re/We0bP348PfbYY1b7Jk+eTLNmzSIiosLCQlIoFPTMM89QRkYG7dmzhwYNGkSMMTEi\noLKykvr06SPa7dRqNcXExNDcuXMdGldCQgIxxmjx4sWUmppKP/zwA3Xu3NmmTb+uA2fBggUUHh5O\n+/bto9TUVFqyZAkFBATU8/A/9dRT5OHhQXfeeae4b/r06eTh4UHx8fFWbadNm0bdunWjffv20blz\n5+iRRx6p58h94YUXGnTkEnFnla+vL3l4eNB3331HRNzu6+HhQYwxK5+IvfuTy+X1Ik6ayy+//EKM\nMXrllVcoLS2NfvrpJwoNDaVXXnlFbLNmzRrq0aOH1Xnp6el08uRJev3110kul9PJkyfp5MmTorP6\nzJkztGLFCkpISKDs7Gw6duwYPfHEE+Th4VHPBkvEIzj69+8v2oMrKyspLCyMVq9eTQkJCdSuXTu7\ntu/09HR68cUX6ffff6esrCw6fPgwxcXFib4Zi03fMn8sxMbG0quvvkpERKdOnSLGGC1dupQyMzPp\nxx9/pB49elj5Cux9LxZH7s0330wHDx6k06dP09SpU6ldu3ai4y85OZn8/PzooYceoqNHj1JmZibt\n27eP5s6dSxcuXGj0e2rOnF+xYgWFhobS2bNnqbCwkKqqqoioviN3xowZoiM3JSXFpiO3uWi1WoqJ\niaE77riDkpKSaN++fdSpU6d6/o4ePXrQmjVrxO1jx46Rh4cHLV68mFJSUmjTpk2kVCpFXxwR0fr1\n62nLli2UnJxMmZmZtG3bNurVqxcNGDBAjAITBIGGDRtGPXr0oIMHD1JKSgo99thjFBISQgUFBUTE\nP+u1a9dSUlISZWdn0x9//EFTp04lf3//Bp3UtbluhD4R0ZYtW6hr166kUCiof//+dODAASthNGfO\nHOrcuTOVlpaK5xw6dIg8PDxo+/btDo3tv//9L3Xp0oW8vLxoyJAhtHXrVoeEvkajoRkzZpC/vz+p\nVCp69tln6eWXX64n9L/55hsCQO+99564b/Xq1QSAvvnmG6u2RUVFNGPGDPLx8aHQ0FBatGhRk0M2\nLUyaNIkYY6RWq8V9AwcOJD8/P9Lr9eK+ayn0iYi+/fZbMdSwW7dutHz5citnnCXsrzajR4+uFwpY\nW0Cmp6fT5MmTKTw8nDw9PSkqKoruuusu+uuvv+pdX6PRUKdOnSgxMdFq/+7du6lbt24UHBxMCxcu\ntDv+K1eu0N13303R0dHk5eVFkZGRFB8fL4YFOiL0iYg++OADiomJIaVSSSNHjqTffvvNYaEvl8tp\n586d1LNnT/Ly8qLBgwfTiRMnrNqdPn2apk2bRkFBQaRUKik2Npaeeuopq/lgj+bMebVaTbfddhsF\nBAQ0GLJZWlrq9pBNIqLz58/TxIkTydvbm1QqFc2ePZvKy8ut2gCw+m6IuGLQt29f8vLyog4dOlg5\n5ol4ZGH//v3J39+fvL29qXv37vTSSy9RUVGRVbuCggJ6+OGHKTAwkIKCgmjy5Ml09uxZ8fiRI0do\n9OjRpFKpxGvNmjWLzp075/A9MvNNSEhIXIds2rQJ8fHxMBqNLT0UiVZCq3XkSkhISEi4Hkno16Ih\nr/+yZctaengSEi1CQ1ExljQDEm0HybxTi4yMDLvHVCoVVCrVNRyNhETrIDs7225UTEBAgM3IK4nW\niyT0JSQkJG4gJPOOhISExA2EJPQlJCQkbiBa5YrcK1eutPQQJCQkJNoUDa02rk2jQn/t2rVITExE\nYGCgmPiqNrm5uVi7dq24VHjatGnisaSkJGzcuBGCIGD8+PGYPn16E25BQkJCQsLVNGreGTNmDBYv\nXmz3uJ+fH5544gmrkl4Azyvx2WefYfHixXj//ffx559/4vLly80fsYSEhISE0zQq9OPi4uqlXK2N\nJVNh3YRTGRkZaNeuHSIiIuDh4YERI0bg+PHjzR+xhISEhITTuM2mX1xcjJCQEHE7JCSkSbm5JSQk\n3A+ZszMKguDSDJsS7oGIIJPJoFQqnf6+WoUjd8+ePWI+8uXLlyM0NLSFRyQhcWOgVquhVCpbrIC6\nRNMxGAyQyWRWSnVTcJvQV6lUUKvV4rZarba7onXChAmYMGGCuF1UVOSuYUlISNRCp9PB19dXSsjW\nhmCMoby8vF6tEEejd9wWpx8bG4u8vDwUFBTAaDTi8OHDGDRokLsuJ3GDQkQgoeEiHBL2kUw6bZPm\nfG+NpmFYuXIlkpOTodVqERgYiJkzZ4pawaRJk6DRaLBo0SJUVlaCMQalUon33nsPPj4+SExMxOef\nfw5BEDB27FiHKzBJcfoSjiJ8/TGoMA/yea/bPE7ZGUC5Fqz3AJvHb3QqKirg4+PT0sOQaCK2vjdH\nNf1WmXtHEvoSjmJ6Zwlw6QJkq/5bT/uhtHMQVr0GBAZDvmxdywywlXM9CP3Nmzdj9OjRaNeuHQDg\n6NGjWLRoETw8PLBt2zaxxGRtcnJy8Nhjj2Hfvn04deoUvvvuOyxdutTt4zx9+jTefPPNZvfVHKHf\nKhy5EhJOoysHKiuA8jLAPxAAQJUVoC0bQIf3ASYjUK5t4UFKuAuTyYQtW7agZ8+eotD/4Ycf8Nxz\nz+Hee+91qI9+/fqhX79+7hxmq0IS+hJtm4py/n9BXo3Q/+ojUMLvYKMm8+0Dv4FMJrBGipff6Ajf\nrgflXHRpn6x9Z8geeKrBNjk5OZg1axb69u2LM2fOoHv37li9ejUSEhKwdOlSmEwm9OvXD2+99RYU\nCgWGDh2KadOm4dChQ5g9ezZOnTqF5557DkqlEg8++CC2b9+OgwcPYv/+/VizZg3eeOMN7N+/H4wx\nzJ07F3fddZfV9Q8fPoyPP/4YX3zxBUpKSrBgwQJcunQJSqUSK1asQFxcXP3PShAwfPhw7Nq1C4GB\nfN6NHDkSW7duxcmTJ7F69Wro9XoEBwfjgw8+QFhYmNX58+bNw4QJE3DnnXcCALp16yaGtH/00Uf4\n+eefodfrMWXKFCxcuNDpz98WUsI1ibaNWejTVW4SpBN/go4dBJv6AGSzngHatTe307XUCCUc4MKF\nC3jsscdw8OBB+Pv745NPPsHzzz+Pjz76CHv37oXRaMQXX3whtg8ODsbOnTtx7733ol+/fvjggw+w\ne/duPPnkk5g4cSJefvllfPDBB/j1119x7tw57N69G99++y3eeOMNXL161e443n33XfTp0wd79uzB\nokWL8I9//MNmO5lMhsmTJ2PHjh0AgMTERMTExCAsLAxDhgzBzz//jF27duGuu+7C2rVrHf4cDh48\niIsXL+KXX37Brl27cPr0aRw5csTh8x1B0vQl2ixkMgFVlXyjMA8AIOz9GYhsD3bbDL7f17yaXKcF\n/ANaYJRth8Y0cncSFRWFwYMHAwDuuecerFy5Eh06dEBsbCwAYMaMGfj888/x1FN8jLVzfDXEsWPH\nMH36dMjlcoSFhWHYsGE4deoUevXqZbf9+vXrAQC33HILSkpKoNVq4e/vX6/t1KlTsXLlStx///34\n6aefxDHl5eXhmWeeQUFBAfR6PTp06ODw53Dw4EEcPHgQkyZNAsBt9xcvXsSwYcMc7qMxJKEv0Xap\nrb0X5IH01cDFNLBxU0VTDvP1BwGATgvh2/VgfQeDxfVvkeFK2KeuEz4wMBAlJSV227cG5/OgQYOQ\nlZUFtVqNnTt3im8F//rXvzB79mxMmjQJhw8fxnvvvVfvXA8PDwiCAICbiiyVyYgIzz33HB555BG3\njVsy70i0XSz2fABUkAdcTAeMRrDuvWvaWDR9TTFo788Qdv/k0iGQyQQqsy+cJBwjNzcXCQkJAICt\nW7eib9++yMnJwcWL3Mfw/fff29V2fX19UV5ebvPY0KFDsW3bNphMJqjVahw9ehT9+9t/6A8dOhQ/\n/PADAG7rV6lUNrV8gD+opkyZgtdeew3dunUTF5+WlZWJTuUtW7bYPDcmJgZnzpwBAOzatUsU+mPG\njMHmzZuh03GFJi8vz+WLVSVNX6LtYhH6qjCu6aedBRgDutZyvPnyHyxducS3U8+A9NVgXopmXZoM\nBsDDA8JHbwEppyB7ZRVYhGMhcxL1iY2Nxeeff44FCxage/fuWLp0KQYOHIinn35adOTa035nzpyJ\nRYsWQalUYtu2bVbHbrvtNpw4cQITJ04EYwxLlixBeHg4cnJybPY1f/58LFiwABMmTIBSqcTKlSsb\nHPe0adNw++234/333xf3LViwAE8//TQCAwMxcuRIm9eaNWsWnnjiCUyYMAFjx44V31xGjx6N9PR0\n0VTk4+ODNWvWuDQ1jRSnL9FmobOJPA5/4Agg8TAQ2R6Qe0D+6qqaNuVlEJ5/GOzmkaATfwIAZHNf\nBbvpZuevm58L4dX/Azp2BS6mATIZ0Lk7ZC++BSZrWxFCrSFOv3bMvIRjNCdOXzLvSLRZyKzpy4aM\nApTeQF4OWI8+1o18fHlbi6YPgM4kNO/CuVmAIABZ6UCvfmCP/wO4cB6U8Gfz+pWQuAZI5h2JtovF\nvNO1F2QrNoJOHQOLs15kw2RyLvgLzG+PnbuDziU267JUwm2sstc/BELDAbkH6McvQUcPAkNGNavv\nG5H27du3ai1/8+bN+PTTT632DR48GMuWLWuhETUPSehLtF10ZqHv4wfm6Qk2bIztdr7+PNJH7gE2\n6Ba+Wre4CEzlpJ20uAjw9ALaRYtRJ2zwraC920DlZWB+Umjo9cT999+P+++/v6WH4TIk845E26VC\nB3h5gTWWC97HHMETpALreRMAcKevs5SogeBQqzBDNmQUYDKBEg8736+ExDVAEvoSbRIymbh5x8d+\nKU8RcwQPAoOBmE6Aty/QDKFPJUVAcJ0CFh26AIEqID3F6X4lJK4FktCXaHPQiT8hzH0AlJvtkNBn\nvrU0fZkc6N4blHrG+QGU1DcNMca4M9kkFSORaN1IQl+iTUFEEH7dAphX3zZF02eBfPEM696Hx/WX\nqBs6y+qalHOxpmCLphgItuEPkMmkgi4SrR5J6Eu0LdLPAZcya7Z9HRH6NZo+wDM/AqiJ6GmMrHQI\n//4HT9VcquHhmnaEPkyCY31KuIXDhw/j0Ucfdbj92bNnsXfvXpeOYfPmzViyZIlL+3QlktCXaFPQ\n/l8BP39gAF+Sz8xx+A0i2vTNNZotKZjLSh27ZhHPykjbvhYfFDYjf+RyQNL0Wwxn6vyeO3euVYeL\nugMpZFOizUBGI+hcItigW4AuPUAnjzho3uFtmFnTRwAX+ih3TOijTMP/Ly6C8IM5va9NTb9tC/1P\nE67iYkmVS/vsHKxE/KCIBttY8ukPHDgQCQkJ6N+/P2bOnIl3330XRUVF+OCDDwAAr7zyCqqrq8WS\nrF27dsXmzZvx22+/QafTQRAELFiwQOw3KSkJL774ItatW4fw8HC8/PLLSE1NhcFgwIIFCzB27Fi8\n8847qKqqwrFjx/Dcc8/Vy7XflvPm20PS9CXaDhfOA5UVYH1uButlXoTlZzsZVm1YTGeu3UebU9z6\nmuPotU2i/KP/AAAgAElEQVQQ+jIZMHA4kJnK99nT9E1tV+i3JFlZWXj66adx6NAhZGRkYOvWrdi6\ndSteeeUVrFmzBl27dsWPP/6IXbt2YeHChfjPf/4jnnvmzBmsW7cO33//vbjv+PHjWLRoETZu3IhO\nnTph1apVGDlyJH755Rds2bIFS5cuhdFoxMKFCzFt2jTs3r27nsAH2nbefHtImr5Em4HOJAByD576\nwNsHsjkvAbG286LXhnXoAvl7X9Zsy+Xc5NMUoe8fBNnj/4CQewkoLbb9hiGTcXt/G6UxjdydtG/f\nXsxx3717d9xyyy1gjKFnz57IyclBWVkZ5s2bh4sXL4IxJmalBIBRo0YhODhY3M7IyMBLL72Eb775\nRsx2eejQIezevRsff/wxAKC6uhq5ubkOja2t5s23hyT0JdoMdPYE0LUXmDdPNMVuHul8Z/6BIAeF\nPpWWAAGB/EGz4A2gML9e/ncA3LwjafpOoVDUZD2VyWTw8vIS/zaZTHj77bcxYsQIfPbZZ8jJycF9\n990ntq+beCw8PBzV1dU4e/asKPSJCOvWrUPXrl2t2iYmNp6So63mzbeHZN6RaBMIf+4FcrPB+g9x\nTYf+AYC2zLG2ZRq+sAsACw6xztdfG7kcoLar6bdmtFqtKMD/97//Ndg2ICAAX3zxBZYvX47Dh/kK\n6dGjR2Pjxo2wJBU+e5YvzvPz87Obi99CW82bb49Ghf7atWsRHx9v5SCpDRFhw4YN+Pvf/46FCxci\nM7MmnO6rr77C/Pnz8fzzz2PDhg1ohVmcJdoAdDEd9OWH3Kwz5g7XdOof5Lh5R6sB8w9qvJ2k6buN\nZ555Bm+99RYmTZrkUJROWFgYPv/8cyxZsgSJiYmYN28eDAaDmL9+xYoVAIARI0YgPT0dEydOxE8/\n2S+wM23aNPzwww+YOnWquM+SN3/KlCnig6Aus2bNwl9//YUJEybgxIkTVnnzp0+fjmnTpmH8+PGY\nPXt2ow8fV9FoPv3k5GQolUp8+OGHePfdd+sdT0xMxI4dO/DPf/4T6enp2LRpE5YtW4bU1FR89dVX\neP311wHwV6GHHnoIvXvb0ZJqIeXTl6iN6b1/AbnZkC1dC+ZItI4DCF+tBZ04DPn7XzXYjoggPHsv\n2PhpkN33eMPj/OANoKQI8n81XHijNdEa8ulLNB235tOPi4uDn5/9H1pCQgJGjRoFxhi6d+8OnU6H\nkpISMMag1+thNBphMBhgMpnEkCcJCUehjBQg5RTY5HtcJvAB8GgenbbxFbQVOsBoBAIc0PSZTNL0\nJVo9zXbkFhcXW5XyCgkJQXFxMbp3747evXtj9uzZICJMmTIFMTExNvvYs2cP9uzZAwBYvny5S0uD\nSbRtNJt2QR8QhLB7ZoEpvV3Wb0VkNLRECFF4gfn4ovLADsj8/OHVbwhktRZ8GavKoQbgH9MB3o3M\nS423N4yMtan5e/XqVXh4SPEcAPDf//4X69evt9o3ZMgQLF++vIVGZB+FQuH0PHPbt52fn4/c3Fwx\nRGrp0qVISUkRw7JqM2HCBEyYMEHcvlYODYnWj+lKDhDTGepyHVCuc1m/AuNlDdVZF0Gnj4HMi67Y\n2Nshe2iO2I6yswAA5UwOXSPzUjAaQXp9m5q/1dXVkMvbVolHdzFjxgzMmDGj3n5nVvq6m+rq6nrz\n7JqVS1SpVFYXV6vVUKlUOHbsGLp16walUgmlUokBAwYgLS2tuZeTaGNQcRHIsqLVGXTamiyZLoSZ\nUzGgKB+080egz0Cgx02gtHNW7cSxO2LeaYNpGKTgirZJc763Zgv9QYMG4dChQyAipKWlwcfHB8HB\nwQgNDUVKSgpMJhOMRiOSk5MRHR3d3MtJtDGET9+B8M4Snv/eGXTlNblzXIlZ6AtbvwZ0WsimP8wL\nrFy5BNLViqIoK+H/OyL022AaBplM1io1WQn7GI1GyGTOi+5GzTsrV65EcnIytFot5syZg5kzZ4qT\nZNKkSRgwYAASExMxd+5ceHl54dlnnwUADBs2DGfPnhXzSfTv3x+DBg1yeqASbRRNMVCYD/p9J9iY\n25t0KgkCL5TiBk0f/uZUDJcvgg2+FaxjV6BCxzWozPPATea5aknB4MiDRy5vc1k2lUolqqqqUF1d\nbXvBmUSrgoggk8mgVCqd7qNRoT9v3rwGjzPGEB8fX2+/TCbD7NmznR6YxHVCZQUAgH76BnTrZJ4C\noSnnEjmUX6fJ+AYA5sInbObf+L4uPXhO/IwUsNpC3z8IzBHNSiZrc5o+Ywze3q5zkEu0fiS3vYTb\nICIuuL28gPIyoFIHNKVouM68YtbH9UKfyeVgt0wEevUXs28yhRJo3wWUkSy2ozJNTVbOxpDJ23Tu\nHYkbAykNg4T7MOh5+cAAczIsvb5p55tt68wdNn0Askefg2zwLVb7WGxPICuDm5YAoLTEMXs+0CYd\nuRI3HpLQl3AfZtOOJW8NDE0V+lr+vzts+vaI6cRLMRbl822tBsxRoS+lYZBoA0hCX8J9VJjj6kWh\nX92k08UoGnfY9O3AYsylFC9ncfNUmabmTaUx2qBNX+LGQxL6Eu6jkgt95qx5p9yi6V87oY+oDgBj\noMtZfPyOpmAAzJq+ZNOXaN1IQl/CfYjmHbPQdNa848qcO43AFAogPIoL/dImLMwCALkMIKHGHyAh\nAYDKSkBNnftuRBL6Eu7DrOk77citKAe8fZoW5ukKYjoCl7PE2rhNsukDUgSPhAgZDBBefQ60veEa\nANcSSehLuA0y2/RZoDnXeBNt+tBpr6mWb4HFdAKKroIsztxAB236loeTZNeXsJB6BijXgnIyG297\njZCEvoT7qGPeoSZq+lSubVpcv4tgMZ34orBzJ/mOJmv6ktCX4NDpY/yPgryWHUgtJKEv4T4qdXzV\nqyW5mTM2/WsZrmmhSw8AAJ1OcDwFA8Bt+oDkzJUAwBcn0imz0C/Kdz7/lIuRhL6E+6isALx9AC9z\n0WsnFme5a2FWQ7CAYCCyPVBdCfgHOpaCAZBs+hLW5GYBxUVA1zi+fkNd0NIjAiAJfQl3UqkDvH1r\nhH5TbfoVLaTpAzzjJuC4aQeQzDsSVlDeZQAAGzaG7yhoHWVgJaEv4TaoQsc1fU8vvqMJmj4JAqDT\nXdsY/VqwHn35H00R+hZHbit5jZdoYUp5Wm7WLQ4AQFcloS9xvWM27zCZDPDwaJpNX6cFSGgRRy4A\noHsfAE0I1wQkTV/CmjINVwTaxQBKb6CVCH0py6aE+6jUAaow/renomlC/0oOAIBFtnfDwBqH+QeA\n3fkAWLf65T3tYrH9S0JfAuAFeCxpucMjQa3EvCMJfQn3UVkB5u3D//by4onM6kB5OYAggEV3tN5/\nJZv/Ed3B3aO0i+yuh5p4ghS9I1EDlWrENR4sPAqUndHCI+JI5h0J92Gx6QPcrm9D0xc+WQFh2QJQ\nRor1gdxsvjDLsrCrDcCkxVkStSnT1PiEVKGAprhV1CSWhL6EWyAioKqCR+8AgKdXvcVZpC3lwt1g\nhPDBG6DqqppjuZeA6A5tq4SfTHLkStSirKTGJxQQxJWeqsqWHRMkoS/hLqqreLy6j1noe9mw6aed\nBQCwidO44zafh7gREXAlu57Jp9UjafoSZkgQAG1pTQoPf7PwN+dzakkkoS/hcig/F8LGVXyjtnmn\njk2fUs8ACiXYkNF82xLdoCnmpqGoNib0pcVZEhZ05fyNz6zpM4vwl4S+xPUIJfwOJB4G+g0BixvA\nd3rVt+lT6lmgWxwPaQNqFq/kcicua0EnrlPIWz56h8pKIOz/tVXYjm9oyniMvphhNqD1aPpS9I6E\n66nQAQol5M+9XLPP0wsoKxU3qbwMuHIJbNgYnsM+KERMSiUc2sHbx3S6xgNvJqJNv2U0fSKC8Nn7\nQHISWI8+vCCMRMtgScsdWMumD/5QbmkvVaNCf+3atUhMTERgYCDefffdeseJCBs3bsTJkyehUCjw\n7LPPokuXLgCAoqIifPzxx1Cr1QCAf/7znwgPD3fxLUi0Oiw5d2rBvBTWhSSuXOL72/O5gogoUEEe\nKPEv4OQRsHseA2uBtMrNooUXZ9Efu4HkJL5RcEUS+i0IlVo0fbPQ9wvgyQfbgqY/ZswYTJkyBR9+\n+KHN4ydPnkR+fj5Wr16N9PR0fPrpp1i2bBkA4IMPPsA999yDvn37oqqqqm1FYkg4DVly7tTG08sq\n944lLwnMi69YeCTo5BEIP30NRHcEm3jXtRqu62jhNAx09CAQ1g4ozAcV5LW4RnlDYxHuZvMOk8u5\n4G8LQj8uLg4FBfazwyUkJGDUqFFgjKF79+7Q6XQoKSmBTqeDyWRC3748h4lSqXTdqCVaN7Xj8y14\neVnn3snLARTePH4ZACKigPIyoLwMbNYzYB5t0PLY0itytaVA+87887/aevK335CUlXBFp/bvICAI\n1BaEfmMUFxcjNDRU3A4JCUFxcTHUajV8fX3xzjvvoKCgADfddBNmzZoFmY00tXv27MGePXsAAMuX\nL7fqT6LtoTboIQsMQnCt71EbEIhKo0H8bkuK8iG074iQMJ6moaprD5QCgIcnQiffBZl/C+XcaQaG\nUjWKAfj7+kLZAnO4sKIcitBwGLTtIdMUWX3+EteWUoMe+oAghJnnNwCUhIaDKsqhauHvxW3qlCAI\nSElJwYoVKxAaGor3338fBw4cwLhx4+q1nTBhAiZMmCBuFxUVuWtYEtcAk7YMLCjE6nsUjAKouhqF\nhYVgjMGUnQnWs6/YhpTmbJp9B6G4Wg9Ut705QNoyAIBWo0H5NZ7DJAgQyjSo8vACVGEwpidLv6MW\nRCgvB8nl1r8BpQ/ocrbbvpeoqCiH2jU7ZFOlUlndhFqthkqlgkqlQqdOnRAREQG5XI4hQ4YgM7P1\n1ImUcCOVdsw7JAAmI6iyAtCogahaydQiooCbR0A25d5rO1ZXYnbkUkuYdyor+PoAvwAgPBIoKbJ2\nnEtcU8hkAOR1dOqAIECrafFw2mYL/UGDBuHQoUMgIqSlpcHHxwfBwcHo2rUrKioqUFbGtZ+zZ88i\nJiam2QOWaANUVtSsxLVQO6e+eeUti6yZD8zDA/I5i8A6d79Wo3Q9LZmGoZz/zrjQj+I1fgvzr/04\nJDgmU41j30JgMJ//1S2biqFR887KlSuRnJwMrVaLOXPmYObMmTAajQCASZMmYcCAAUhMTMTcuXPh\n5eWFZ599FgAgk8nwyCOP4N///jeICF26dLEy4Uhcn5DRwBdh1Y3e8eJCX9i0Ckg5xfe1a5m0yW5D\n3oIrcs1Cn/kHAD5+IEAK22xJTKb6mn7tVAxKn/rnXCMaFfrz5s1r8DhjDPHx8TaP9e3bF++8845z\nI5Nom1RW8P/rmnc8zSUTk44BEZFgfW7mZojriZaM06+t6Yfxz5WyL4D1H3btxyIBmIy8cFAtWEAg\nfxiXlfK3sRZCSsMg4Voqdfx/O5o+SAC7ZRJk98c7XnC8rdCCaRioltBnvn5An4GgQztBBoP7r10r\nO6qEGVvmHctvooUzbV5nvzqJFqeCC31Wd0WuxaYPgEVdZ2YdCy2ZhqG2pg9ANvEuoEwDOnbIrZel\nrHQIcx/gxXAkajAZ65t3FOa1Si38kJSEvkSzIJPJOlpFNO/Y0fQBcRXudUdLplbWlgEenjWCpVd/\nILojaP8vbr0spZ3jPoxr4DQmIp6yuC1gS9M3fzckafoSbRlh5asQVv27RvCL5h07Nn2FsqZu7vUG\ns5RLdL3QF/7cA0pOsh8OWl7GTTvmVCeMMbDhY4HsDFDRVZePRyTnIgCAzG947oQO7oDw0t9A5kCS\nVo0tTV/pzf9v4egdSehLOA0RAdkZQPJJCO+/CtOieJAl4ZetOH0AaBdz/dnyLbhJ0ydNMWjTagjv\nvwLatNp2G7PQrw0bMJwfS/zLpeOxum6Oee1NpXuFPhGB9m3n6ztKi916LZdgS9MXhb5k3pFoq1To\nuDknIAhIPQuoC0DH/+DH7MTpX7f2fMB9cfoWIeEXADrxp+1FV+VlQJ3UFSw8EojpDDrpHqFPBj3P\noQSIvhy3cSmz5lrqQvdeyxUYjWB1NX0PT56fqUoS+hJtFTU3G8geehqy97/ktvqKcn6sbhyyl9m8\ncz3HjYvROy62O5uFPOs3mC/uST8HAKC0syBzyUloy8D86ucrYgOHAxfOiykiXMqVSzX36mahT3/t\nq/m7pA2klzAZ62n6jDGeZFAy70i0WSy24tB2YL7+NbnxFd48lWxtQsLBpj4INmzMNR3iNYW5KWTT\notn3vhnw8ACdTQRlJEN4/xUIH7wJ0mnNNn3/+udGmFfnlpfWP9YMhN93gXZt5RtM5n7zTnIS0OMm\nvlHcBjR9W+YdgJt4WtiR2wbz10q0FqjInHI71FwYp0Nn4NjB+vZ8mB2L0x68hqO79jDG+A/dXZq+\nnz+oW2/Q0YOgw/u4Wa1EDfrhS/6GZUvT9/LiC4L0rsvDQ0Sgbz4BjAYuxIJUIMsbnrsoLQHrcRMo\nN6uNCH0bjlyABzJINn2JNou6gAt4c4UrUdO3IfRvGGRy19v0LQusPL3A+g/ly/ijO0A2/w2wIaNA\nh3YAAFh0p/rnWqKmDNX1jzmLXg8YDWBDRkH27GLA19+t5h0yGvhDLTAIUIWB2oJN31YaBgBQKFt8\nMZuk6Us4DRVd5WYbS0U0i9Cv68S9kZDJ3WDeMQtsTy+wMbeB9R4IFmFexn9/PNC5O1ifm2v21aZ2\nojtXodPy/3vcBNarH7DrR6Bc67r+62KprRzAhX6bSCRnT9NXegNVFdd+PLWQNH0J51EXACE1NY+Z\nfwAQHFp/YdaNhFzmcvOOmErBywtMJrcS7sw/ELLxU20LfPM5AGr8Aq6gggt45st9CMzb172O3DJe\nb5YFBIGpwoDipjtyhY2reLK/a0VDNn3JvCPRlqBTx0GXMnmMflEBWGiE1XHZrDmQ3XZfC42uFeAO\n845FS/fwbPq5ZvMOuVLTt2j1vubC9T6+7nXk1q43qwoDKnVNWgxG+ZdBh/eCDu8DOfiWIPyxG8Ku\nH50ZLceOps8USilkU6LtQCf+hPDBUgifr+av+NWVNU5cM6zfELDuvVtohK0Amcx90TuWsNemIGr6\nLrTp68xOW7OmDx+u6burOIhYV9Zi3gGAJoRt0p5tPOMlY6ADv9lvdykTwjcfg/TVoJ+/bbBtg9cT\nBP62Z0vTVyhbPGRTsulLOASpCyFseJ+/nl7KBO34HoAd5+GNjFscuWah7+mMpu96mz7p6mj63n5c\ns9XrAYUTD6bGKOXmHQQEgalCQQCE7zZxs1afgQ2PtaoS9Nc+sKFjQFUVoD92g6Y9yDXu2u20ZRA+\nfIObjrx9eYRQnTYOY/n+bQp9b0nTl2gb0JH9gF4P2bzXAZkMtPNHoEMs0LNvSw+tdSF3hyPXIvSb\no+m70pFr0fTNIaIWx32lm8I2taWA0hvMSwFEdwBiOgPnT0PY/m3j5+ZfBvR6sH5DIJs4HagoB+39\nuV4z2ryeO4z9A0WFBtVVziVHM5lzA3nYceTqq1o0cZwk9CUahHIuggrzQUcOAN3iwGJ7Ar36AQBk\nU++vidyR4Mhkrk+tbBHYtoRIY1geFHpXmne03L9geaBYQnTd5cwt03B7PgCm9IH81VVgI8YDeZcb\nNSnR1Sv8j4goPnf7Dgbt/AGkq3lAUVUlKPEvsFsngo2abO2INzuRm0RDmr5SyRfLtWD9YknoS9iF\nDHoI7yyG8NrfgfzLYEPHAABkUx8EmzQd6De0ZQfYGnGXpu/p5dwD1mzLdm30Tjng61+T0VPU9N0T\nikilJdyeX5vIGD4Orabhk6/m8vs3VxOT3TWL+x+OHazp/8wJwKAHu/kWsGFj+c6Yzvx/jTNC36zp\n21ucBbToqlxJ6F+HUGE+KD25+R2dPs61N4WShwsOGgkAYLE9IZvxpKTl20Imd/2ru8FQY5tvIowx\nfq4LhT7ptDX2fKAmRNetmr610GeWmgx5lxs+9+oVvpbE4g9p35m/pVhWkwPAiT95/916gbWLhmz+\nUsgefMp8bVdr+i2fXlly5F6H0E9fg04dg2zlN/Vz4DQB4a/9QJAKstc/BMrLxLhsiQZwy+IsvdNC\nHwA3w7h0cVa5tdA3r8iminK4RQ0o04D1quM7iozh18zLAbPk5LEBXb3C8w+ZYYwBwSE8RTP42yyd\nSQAbPhbMnCWV9eoH0vIFYVSqafo9iZp+fcc7U3jztBgt6MyVNP3rEMrP5a+POZkQjuwH5WY3vY/y\nMuDsCbAho8F8fHmaXonGkbspeserGULfU+HakM3ysppwTaCWI9f1mj4ZzCkY6pp3gkN5JEwDmj4R\nAVdzwSKi65wbUpOp83IWoK8G69Xfuo2vP/fPuFrTbwUlExsV+mvXrkV8fDwWLFhg8zgRYcOGDfj7\n3/+OhQsXIjMz0+p4RUUF5syZg88++8w1I5ZoECICCrjzio4dAm1YWZMNsSlkpAAmE9iA1m23//6c\nGuuOt6Jl+W6I0ye93rmFWRY8Xa/pMytN343mHW2tGP1aMMaAdtEN1+Yt03Dlp85qZRYUAmh4IRbK\nzuA7O8Zat5HJ+DVLb0Cb/pgxY7B48WK7x0+ePIn8/HysXr0as2fPxqeffmp1fPPmzejVq1fzRyrh\nGFqN6FCjfdsBIlBBXpO7oUsXeMpcSz6dVsqRHC2OXnZzhsem4K4sm84szLLg5WW78IqzVGitNH3m\n6cUfSs0U+iQIoLpRRuaUCywopF57Ftm+YZv+1Vzerq6mHxTCs5MSAdkXeErqkPD65wcE1ywMawpm\nTd+madVSZ6IFbfqNCv24uDj4+fnZPZ6QkIBRo0aBMYbu3btDp9OhpIQ/HTMzM1FaWop+/fq5bsQS\nDXPVLOCDQmpeM82af1OgS5lAu+h6i1haG1d1BmiqjG5bDdpkZDL3mHecWZhlwYWaPumreV91/Tu+\n/tzs05y+9/8CYe6DEL5dzzNrAjX1fcNsmBej2gMatVX4pVV/ly7wP+rmJQoO4WmhdVqu6Xfoajso\nITDYOU3f2ICmrzQXR2/N5p3GKC4uRmhoqLgdEhKC4uJiCIKAL774Ao888khzLyHRBMii3YwYz3eE\ntQPKNKCmZvbLvgBW55W3tVFlFFBaZYJRALT6llvsYkWrdOS60KZfUScFg4UgFQ+tbA6XMgHBBNr7\nc01d3yKz6S60vibOOnblf2Sn1ztGFTrQb98DXXrU0+JZsPmtoTAfuHLJ7jxnAUFO2vQtQr8Bm34L\nOnLdFr2za9cuDBgwACEh9V/L6rJnzx7s2bMHALB8+XKrh4hE09BqNajw8EDoQ39DRYA/5GGRKFvz\nBoIMVfCMsV2qUH/+DAznz8B3+kMAAJOmGEUaNXx79YVvM7+Lj/7Mgr/CAw8PimlWP7a4qK71IFP6\nITSk5bN7liiVoAoTVC6cw2oSIPP1Q7CTfZb4+kEoK0GIE+cbc7NR+t6r8H/iHyBDNfRnTqACQEC7\nKChr9VcS3g5C4VWnriH2oSuD0LErjFnp8NGVwS80FKVaDfSqUIRF1s8iKtw8FIUAfAquwHfURKtj\n2s83o0JbCtW/3oVnWJjVMX3HLigB4J2ZAp3JhICbBljdi4XyyGjo/tqPEJWK2/gdRJ/vhxIAgSoV\nvOr0S4EBKADgK2fN/m05S7OFvkqlQlFRTfIjtVoNlUqFtLQ0pKSkYNeuXaiqqoLRaIRSqcSsWbPq\n9TFhwgRMmDBB3K7dn0TTMGVlAKERKK6sBsbfxc00ADRpKWD+KtvnfLYSuJCKiqFjwTw9QWdPAAAq\nQtqhspnfxc7kfHh7yjClk+vNRKm5Na/1mVeKEEAtm8gKAEwmAaiudukcNlVWAkRO92kCgIoKp84X\nEo+BMtNQ8tpcK7OVViCU1+pP8PYDqc80675NV68A0R2BkiJUXLqIqqIimC5nA6pw+/1GRKP8XBIq\nx9xh3dfxP4C4/igNCgPqnEsyLvZ0B3jxGa2qndW9iPfkqQAEE4qyMrnW7yBUzMNBS8t1YLbGLfeA\nrljd7N9WXaKi7KTXrkOzhf6gQYOwY8cOjBw5Eunp6fDx8UFwcDDmzp0rtjlw4AAuXLhgU+BLuJir\nV4DajitzqCVdvWIz3pgKrvBIHYDXvI2MAWWbbaHtOzdrKCaBoK40glUCBhPBU+7aKO6r5Qbxb02V\n0aV9O407bPr6ajBn8u6YYZ7NcOSW8igX9B7IzSAR0aCjB3nepdoEqQBtKchoAHMi0oiIgBI1WJ9B\noBA1SG1ePFV4Fayn/Th81rkbKOUUiMjaLq8pButqJ4AkIJiv0s3PBbr0AAsJs9mMBap4TL2muH7I\naEM0FLIJcBNPC0bvNCr0V65cieTkZGi1WsyZMwczZ86E0eyomDRpEgYMGIDExETMnTsXXl5eePbZ\nZ90+aIkGKLrKqxmZYUpv7pCyE8FDh/fVbBRc4UI/7SwQ1aFmeb2zQ6kwQDD7V/O0enQIcm0GxgKd\nATIGCAQUV7YSoS+XA+Ri/4LR0DxHrpfCeUduqQZQKCH/+79q9g0dXb9dkKqmvR0h2iAVOh67HhwC\nFhoBupjGH1QaNRDazv55nboDRw4AJWpAxc0lpK/m+YFsRPwAAPPwEMMx2fBx9vu22P41aqBDE6LY\nGkq4BrR4cfRGhf68efMaPM4YQ3x8fINtxowZgzFjxjRpYBJNhwx6/sOpWyA7LBJUaEfoHzsEdOwK\nZGfw1YsGPZCeDDZ6SrPHU6irEcTZmmqXC/2r5XpE+XuhUGeAppUIfSaTg9yyOKsZn52nl/OO3FLH\ntFwWZNGK1c4J/RJe95apQvnCwBN/ckcrEQ9GsHfdTl35dbPSRaFvicEXhbYtgkL4eoPBt9pvE2x+\niJSom7YqV9T07YhXbx+QlIZBwiXULW5hhkVEgU4fr/cKTOpCoDAf7IGneEUhi6nHoLd6W3CWAl2N\n+eVSqQtXhJq5Wm5AhJ8njAKhpMrFgtZZ3BG904KLs6hMw98UGyPQoukXO3UdlHA7OIJDudZvMon5\noxA5GAkAACAASURBVFhYhP3zonhwAhXm1Qhmc4oFW7H9FtiIcUBlhfUis7oEBnFzXRPLM1JDi7MA\nrum7KTmdI0hC/3pCFPp1JnJsT+DPPTy3eGR7UHoy6GKaqMGxbnGgiCi+iCvlFDdRuKD6lUXoh/t6\nukXoF+gM6BHqDZ1eQEkr0fQhd61Nn4i4eac5aRi8vACjASQITYpCAcDj1KNtR31ZYTbvkKbYqfw7\nZBGswaFg1ZVcez9/mu9rwLzDvH14wjd1YU1flgdIA0JfNu7ORsfEZHL+MGtClS4ADtj0vZuVsoJy\nLgI+fnZ9EY0hCf3rCXMMdV3thfXsCwJAKafAIttD2LIBuJgG9LiJ50KP6QQWFgnKSOYZFDt3B7Os\nHGwGhToDgr09EKtSIlvjmrhkgQgCATml1SjXC2gfqICmyoQcNzxUnELm4hW5YgGVZubeAXi2zqZW\ntiotceytzy+Aa7YaJzX94iKuVQcFA0Z+z5R0hD9MGjMvqUJBxTVC3yHzjqMEh4DMbw4OY15YZt+8\n4930B0kthI//A+irIPvn27xQfBOREq5dT4hl7OqYd8LaAaERoJTToCuXuMAHgNQzQGxPrtFERPIS\ncZcywYbYcNQ5QUG5AeG+nugQ5IX8cgMMLigusuqvPCzckYXvz6mhkDOM7hSAYG85SlpT9I4rzTsG\nswBpbpZNoMl2fdJXc43UAfMOk8l4O2eFfkkREKTic1EVxqNrjEawKfc1/naiCuNz14JGzSNkvJuv\nuCA4xOWaPmuuI1ejBjTFED5c5tTpktC/jhCXo/vUt1Oynn2BtDOggzv4ZDTXFmVd43iD8Cjxf3br\nJJeMp0BnQLivB4KVHhAI0DVz1SwRIfGKDhdLqvF7thZjOgfCTyFHsLcHdHoBegceKgIRjIIbUza4\nOsumRVA3S9OvqZNL1VUQDu9zLG2FZYWtIzZ9gK/K1aidy+paUiQ6TpmHJzfNBKnARjU+F5kq1Fow\nl6iBoBCX1HtgwaE1eXocpTFHrtLHaaFP1VW8CpqPL3DpglPpHCShfz1hR9MHwGvZVuh4Era+gyGb\n/givO9p3MADzknaZDLIZT/CQtmYiEKGowoAwX094e/JpVmFontDP1epRVm1CV5USXnKGO3tyYRSs\n5ONtzK5vEgjLDl7G4t1NF0oO42pHrks1fT3PvLpxJZCV0fh55mRjLND2or56BKmAlFMQXvs7KOGP\npo2xRM0FrBnZA/GQxS/kydwaQxUGlGtFAUgadU0IqZOsT7iKt//I5Zp+dVXTbPANpWEAxJBNp/JF\nmfP8I7oj/9+JhHCSTf96QlfOzQs2XmvZwBHAjGJA4Q3WfyhYYDBkq7+tKXkXGQPZ6s1gTbX52iGz\nuBpGgTtxfVwk9FMKuHb0/IhIhPp6QunB+w3z5ZEthTojIvzsC4nNZ4twPFcHXy836joyucM1coWf\nvgb8AiAbP9V+I33zbfrMU8Edo4ZqwBy6SzmZYJ27NXyiqOk7tjCJte8CSk4CFEoIO3+E7OaRDmnb\nfGFWIdBvcE1fA0c4dE0AXOgD3C8QGcMfIM0IRBCIcCirjI+rk/lBVKK2+QZtk0Y1fW++lkNfXZOL\nx1HMQp9Fd+LRTaUlDYa02kLS9K8nKsoBH1+bPzTm6QnZpLshGz0FzPy6XredqwR+hcGE9w5fQbBS\njhEd/OHjKRf3N4eUwkr4K+SIDvASBT4ARPhxoZ9fbnYAEiEpT4clu7Pxwo4s/HBODaNA+P5cMeSM\nm5lM7jLxyGWAvgqmFYt4MZsGoAO/gbZvbjiu3+zUZM1anFVj3kGBOYFZzkU+hvxcmFa9ZjMhn5hA\nLcAx8w67/T5ere3O+3ncvKMlO3VaPjYnHa+iM7O4kJeqLC1uMHKnMS5pqlFWbYJWL6Dc39xPU+z6\njmj6gHMmHhdo+pLQv57QaQGfli9puOdCKXLL9FhwSxQClR6ieafS2HRN3yQQDCYBRISUwkr0CvOu\n97AK8/WEjPG4/V9SSzDru3S8ui8HBToDyqpN+C5ZjXytHkaBEKvimpVW33wTTOKVcuy9oIG6omY9\nAuvWB4juBGSmgfb/Yvdc0pbyVMTlZTWhibawRO80q3JWLfOOOVUx5fCcTHTiT+BsIpB7qf55ZSW8\npkJAoEOXYTI5mFwONnw8oPS2Kj7eIOYQSxbsXAiiZTEYFRfyMZtMzRL6p/JrHoD5XvwtRwwDdYTG\nQjYtkXFOxOqTlqevZjFc6JMTWUAl8851BNWtXdpCFOkMUHow3BTB0zhYzDuVTph31hzJw7HL5Wgf\nqMAVrR539qivdXrIGEJ9PJFfbsCJKzoEKuT428Bw3NopALszSrEu4SrOXOU/sJ5h3khTV6Gs2oQg\npfPTXyDCit+voNIoIEgpx/rpsfCSy8D6DYa832AI694GHT0Iuu8J21r6lRohS8d/B+s9wPaFLOYd\nD9c4clFo1vQvZ4EEE1+vAdjOG19aAvgHiLVjHYUpFDxazNFIHjFG30lBHaji0T7FhaDUs3wMXbo7\n1xeA0/k6eMkZ9CZCHnkjlsmso4Maw2QE5B52TVtMaa6T68yqXEslsagO/J4lTf8GR6e17cS9xpRW\nmxCgqBEUzXHkXirVQyCgsMKA+JvDMaWbbftyOz9P5Jbpka2pxpAYf4yPDYKXXIb2gVzgHcnhTu6e\nYfzVWlvdPE0/t0yPSqOAYe39oKky4XS+tdbGho/j38eZ4+I+SjsLupzF/75iLvPXsy/o5F8ge85f\niyO3WYuzuNmONGpuAozqwJ2TBXlAZio/ZkN40MV0ILK9c9cMCHJYIIn1alXOpRpmHh5c8KsLgXOJ\nvBKWk7UgjALhXEElbunIf0d5FSYgWMWTETqKyWRfywdqfG5OmXfK+EPc25f/1ksloX9jU1He8LLy\na0RZlQkBihotujmO3NIqI4Z38MOGu7tiak8V5DLb2lM7f09kFlfBKBC6BNf4JjoE8r/PXK1AoFKO\nSLOjt6yZQj9dzSNF7v9/9s48To66zP/vqr57+pjuua9kct93uCFcAcVbcNlFRTxYfi4Ki8euirru\nggeesKKouyq6uKgsi7u6isRwQ4AEkpCQezKZZO7M2fdd398f1dXTPdMz0z13SL1fL15kZrqrv9Vd\n/dTzfY7Ps7ocm1HmpfRNJcPK9eAqRby2A1BHASo/ugflN/+u/r3zFNjsSBsvUGUHRps6VWTJZlcg\nzo93dpFIZeUstOemdxdaklTseWUoRjwsTCBCAWhvGVPhciwkl6dwL3SgVzWSxShZDqemHnFoL+LA\nHqQV64venWgcPB0mklQ4r95Jmd1IZyCu7lqKMvrJ0ZO4MBTTn4gUQ2AQnG51F+Ge2DhH3ei/mZgj\nnr4/lsJtHfrSmWQJo1x8eEcIoR7LMn4YpsphRjNzC71DFRFuqwGnWSYloM5pxpnegUzW02/qi2A1\nyswvtXBOnYNdbUH+641eDvWoX2TJYID5i4c8+tYTqmFvPoJIJNTf1zSoxhFGNZCiyJLNXe1BHj82\nyNHeLC8yvUsQmtFfd44q+vXnR4ceM/z1j7wBQiAtW1vQ644grWJZUFlify+4vRM21ADyuz+gNob5\nB2HVxgkfZ2d7EJMssb6mhFqnmc5AAqm8ukhPPzm2p29Rjb6YgKcvAn5wpnMsE5zspRv9NwlCSame\nQ6FlZdOIP5bMGFdQq4RsJkPR1TuRpEI8JXBZxzcG1ekKHotBotY5ZCAlSaIh7e3XucyZsNNkPf2j\nfVEWey0YZInzGxz4Yil+9Xovvz889CWUquugu10d+H1wj/rLRBxONkFnqzrYW/NuR9umF+np+9LC\ncwd6srxITYZBS9ZW1yHd+Al1h2E2Q3U9YtjriyP71b+NV9Y5Gu5SVY6gAG9WDPROOLSjIS1ajrTl\nLSDLSKvWT+gYQgh2tgVZV23HapSpcZroSnv6DPYXPpMglRrb08+Edybi6fsyRl8qIoSWjW703yxE\nwqoM7VwI78RSuC25htpukosO72gGrJCEq1a22eixjAgBZRt9i1HGbJAm5eknUoITAzEWl6ke23kN\nTv52cyX1LnPuzaS6TjXy/T1q/Xp6Vqt47UX1y1rbkKmBH3W+bCxt9AuM6ftiarmg1tMADOnt+PrV\n0IDVjnzOJUjvfj/SpdeoCdTh4Z0j+2HRigkNRAGGbmaFGKWB3pzGrIki3XAL8j99f0x1zbFo9cXp\nDiY4p179DtU4zPhiKULedB28NthlPJLjePqTLNmUnC6SiiDgqgB/gbupLHSj/2ZhrG7cGSSWVIgm\nRU5MH1SjX2x4RzOgw28g+ahOx+oXekY2u2jJ3DqX+n+nxTApT7/VFyOpCBanw0hGWeIdy7zUucwE\nsiSeper0XOBTx6HpINKmC1WV0+2/B4MBKR33B4aqMoZz4qjaXVrgDk67UR7ujWR6ESSzBemGW6Bh\nAdK6czOPld/xN8jXfyzjMYq2FnUKVTymDgwfbfJUAWTGC44TftAmZjEVRt9oQipEEXQUdneqMiab\n69JGP71j7CpJD1YvNMQzXiLXbFFLYYs0+kIICKqe/u8P9XNbciMiHi+6Ckgv2XyzEMqvsDnTaMZ0\neEjGZlQ9/X99qROLQeLj547fRaiNQCwkvOMwy9y0voJNdSPPf0NtCUtbrCwvVz0sl8UwKU+/M90E\nVu/O9b5dFkNuLD1t9JUn/6CKh63aCLEoorMV6boPI9UvUL/IZkvekkmhpBAH9yKtPadgHRnN6IcT\nCrf/8QQLPVY+c3GtKiU8mpyw2wP+AVV9tf0U8h3/rO4aqycxzH6cXEWGYEDdDU0yvDMV7O8KU+s0\nU25Xdzc1znTTn8nNQkD0dhcmGz1OIleSpIlNz4pF1bJbp5sTAzF8wkjYYMXpGxyq/S8A3ei/WfCn\nqzBm2NOPJhUMEpgM6qZxNO/cZpIZjKY4cDqMoUAD5i8ivCNJEteuyr+tr3dZ+PZbGzM/T9bTP52e\nzavJP2i40sfNDKtxulUP/egB9d/L1iCVVaqjKC9/e2bduD35Y/qnmtUd3Gg1/HnwxZIsL7dxuDdC\nmz9eWEjNVaoak6aDqihbunZfyp61XCxpT1/4B8c2lOn696kI70yGlCI42BPhkvlDU+eq055+p2JW\nh9gU6OmL8Tx9SBv9ImP6WqWV00132vEImOw4/YNQVdhQdNDDO28aRGuz2qwx0brqCfKl7ad4YGdX\n5mef5p3niemH4in6wgm6g/GCZBA0r3X4sSbLZD39nlCCEpOMw5y7LrfVSEoMlaZKkqRqwQDSxgvU\nbtUq1evO8dxdpYg84R1xQE3+FjPFzBdNsaTMypcurWfrIjcDkWRu+WY+NK883QgmXt+p/lxVU/Dr\njsDhVHWgfIOIrrZR486iqy39WpO4wUwBzQNRwgmFNVVDHrPVKOO1GekMJqC8svCyzfFKNkHtWI4U\nGd5J91RI3oqM4xEwlRSt968b/TcJouUYVNVNeph5MfSEEhzri/JG95DHMhTeGR7TN9ATSpBUICWG\npmr95+s93Lejg1AeWQRfLInVKGMxTu1l6jQbCMQmrr9/OpSg0jEywenMUxkkVavGbMxZrOnyxmyE\nEIg9L6tx+ALr1+MphXBCwW01cE69gxUVaudnb5ZMRD5GHP/gXnB7JjVIR5IN4HQjXn0e5cu3qk1T\n+ehuV+PblZO4wUwBWsf26qrcc65xmugMJKCiBpoOq3OkxyOVGn0oukaR4R2hKIjHH4WaBmKLVmXG\ng/pNJep7WAS60X8TIISAE0fHV02cYvZ0qnKzp0PJjIefMfp5wjvZDmeHP044keJ3B/t5+oSfz207\nOcIj9UVz6/2nCpfVQHASomung0kqS0Ya/XzloNKGC2DtObBk5ajHy1t6t/cVONmEdNk1Ba9L2xm5\n0zfcyoz66NhGP6OiabOr+YVEvKhwwag4S9WuX0Ds25X/MV3tUF45OUG5KeDg6TD1LjMeW66xrnGa\n6QzEkd9+PaQSKF//zPg6POPV6YNq9IclYMVgH6m7PzW0+8nm9Z3QfhLpbX9FT2To+vJ7qqFTN/pn\nH/29qtFYMHG9kUIYHKZXv7sjhFYdebxf7VD1RVPIEiPki7WuXI2OQJyXW4MkFMEVC120+uI5g9TV\nYyULqtwpFqfZgIC8u4vxEELQHUqMbfSzK3jWnYvhti+P3Xjk8kDQj0iq769QUiiP/RKq65Euuqrg\ntQ3Pp1QUavS18E7jkkx4UKqcAqOfJcmshaqGIzrbJpcwniJODsbzVn7VOM0MRlNE5i1B/sd7IBJG\nPPv42AcrJLxjs4/sYWg9oQ5GeeqPiM42lJefBtRrTvnjI1BRjXTOJZnQDkCgtBox1Z7+Aw88wM03\n38xnPvOZvH8XQvDzn/+c2267jc9+9rM0N6vqfS0tLXzxi1/k05/+NJ/97GfZsWNHUQsDEH09pL79\nhQkpyZ1VtBwDQGqcPqN/YiDKR37XxHMtqlxAShHs6wpxQYMTiSFZgkBad0celqzNNvpGWaIjEOfZ\nFj/VDhOXNqrNJsOHoPhi0+PpZ8IwEzD6gbhCNKnkDe9oa/UXGzrSJlNpibruDuhqR7r6PWpnb4Fo\nuy3N0y+3G5FQ5wyMicOpSnIvXa02jMGUePpa2Eg6/zI43Yno6cr5u1AU6G7PhMDG45W2AF/YdnLK\nZbGjSYXToUSmtDebTAVPIKG+N2s2I57fhkiOcSMtIJErWUaGdzQFTfHKMyg/+CriZ/eiPL9NDY2d\nbEK65n1IBgPdWTfxgKscxsiZ5GNco3/ZZZdx5513jvr3PXv20NXVxfe//31uueUWfvrTnwJgNpv5\n5Cc/yfe+9z3uvPNOfvGLXxAKFTkB/sQROHoAcWBvcc87yxDNR1TPor5x2l7j+RY/ioBH31BHx50c\njBFKKJzf4KTOZaZJ8/RjybyJV010zWqUmF9qYV9XmH1dIS6Z78JrV41U/3CjH01lDNhUoh3TF03R\nE0rkllmOg+Zl5fP088X0C0Ea3siUFiCTijS8g5nwjroOk0Gm1GakZ7yYvmxA/pcfIL31WrVhjKnx\n9KW15yCduwXp7dcDebz9/h41lFSgp//SqQAHeyJTPg+5zacmsLUmvmw0rabOgPoY+fK3qT0Nr4wh\nG12Ip1/iGOqt0dB+DofgdAfUzkP8549R/u074K1AuuByQJUQN8kSbouBgK1ULeUsQvp5XKO/cuVK\nHI7Ra79fffVVtmzZgiRJLF26lFAoxMDAALW1tdTUqMkZr9eL2+3G7x9FVGoURDD9JmjyrzojEMmE\nqlu+bM20xUWFEOxoDWA3yZz0xXitIzRUq+4ys9hr5eDpMA/v6+F4X3REEheGjH653USd00ybP47V\nKPP2ZR68tpHjDlXdnekJ73jSRnEgkuThfb18+clTBTeOnQ6p553P6NuMMkZZoiec5N92dY0Ih43K\nsEamTMy4yDLGIU9/6D2rLDGOCJvlQyotU5ublq5WJR/mLy7qtfMec/PFyH/7WbUyp6oO8eTvEdEI\nouWYqimUHjIjFWj0mwfU7uTe8XYuRdLqU4+b39NXf9eeNvqs3AALliJ+/W+IU835D1hIyabLo/Zs\nZHv7Qb9a8VTTgHTxVcj/8HWkS65GWrMJ+UOfzHRHnw6pY0jdVgN+c7pwo4gQz6TdqP7+fsrLhy7O\nsrIy+vv78XiGdM+bmppIJpNUVVXlPcb27dvZvn07APfcc0/meCEUgoCh7QRl5bPfvDEXiTzzZ/yD\n/ZTe9iUs0/QeNfWE6Awk+NRlC3loVxsvtEVYXqX2A6xqrCZmsLG76xi/3d+Hw2zg4kUVOdcEQHXI\nAHRQU2pncbWT5076ueXCRpY0VCOEwGw4TgRT5nk9QXXcYn25e8SxJouxJAG0EJctDMRDRJOCgz54\ny/LxXyd8SjUQy+dV5725ldpO8MLJAL5okoVVHv5m4/hNaCmW0AuUxKPYy8sJxiKEgPJFS5DMhU8z\ni8sBzAaJhurKTElovbeXI93Bwt/D8osRv36yqLBSIcRu/RyDX7kdcectiIAPQ1Uthqpa4kDZyjXI\n48y0jSUV2nyH1X8brFN6TfQeCWKUJVY31mA0jPSDKx0t9MakzGumvvQd+v/xZsR9X8F1x1ewbDw/\n93gIjPYSSsvLEULw4Yf3cPXySj6waejmFqlrwA94DBLG9HH9yTgxp5vy+x8GSUKSZfj7L41YT1+0\njQavnXhSIZJQP6eS0Tq68zDtzVkDAwPcf//9fOITn0CW828stm7dytatWzM/9/aq21ulR62LTZ44\nSk9X58R1QN7EpH73n1DTgL9hEVJvESPdiuDx/T1IwHqvzMteC0e6/ZhI4rYaiPgHWeGGh65bTEoR\nGd2b3mFrSaZrkt0mwblVJkJrythSa8o8zmsz0N4XyPy8vUm9iBc4xIhjTRYhBEYZWnt9dPvUZNof\n97ezqXz8uobj3QPYTTKxwAC9wZFtRw6TREvaE325uYet88Y32kLIYLERPHKA8KaLUdpPgdNNnz8A\nBMZ9vkbXQBCXxUBf39BW32VQ6ArEON3TMyLPMqPUNiJdcx1ix9NI195E6tUXSB07BItX0pdIjXvt\nHuuLZKq/TnT30+udunM52jVIrdPE4ED+oS+1ThNNp/251+Gn7kL86BsMfvUzyJ++G2n5kBppKh5H\nSSbp7e3FF03S1Bsm+UYnb5k/lCgWcnp3e7IZyaT+PtV7GuwO+gbGzmF2+SI0upz4FUFbLKVeO02H\nKVSYetJG3+v15rwZfX19eL3qXTscDnPPPfdwww03sHTpBJKMWngnmYS2FrW6QCeDCPjg1HGka28q\nuE1/IrzWEWJZuY1Sm5H5pWZebg3gtMQzejcao2ndw1B4p8JuosZp5v1rc0fjeWxG+rNitbvag5Tb\njTSWTs3c3mwkScJjNdIfSdIbTmKUJfZ0hhiMJCm1jf2VaPXFqXeZR32/syUjDvaEc26EwznQHcZm\nklUp6Lp5GeljMdA3oSlSvmhyROK7osREUhH4oqkR5YgzjXztTYj3fkh97665rqjnnkiHdmB6wjv5\nKnc0Gtxmth0LowiRuXFKVbXIX/g2yr/cjvLQD9Wqp4Fe5Bs/kRPT1/IFLYMx+sIJytISD5nkfXYn\ndjAAzqGO4HwkUgq+WIqydB7MH0upjWNFzPCddMnm5s2bee655xBCcPToUex2Ox6Ph2QyyXe+8x22\nbNnC+eefP/6B8iBCAbWygPQUH51cTjYBkxsNNx6DkSTH+qJsqlNjh/NLLQjgaG+EmjwVLKPhsRox\nGyQaPfmNuNdmpD+sfpnjKYW9nSHOqXNM283MYzPS5osTTwm2NDpRhHqjGY9WX4x5Y9yItCT2kjIr\nobjCycFY3scpQvCtF9p5cLeq3CjVzYf2lrQAWe+EBMja/SNvxKU2dT2+KU5+TpSJfp7H+6OUmGTq\nXOZxm82KIZ5S6A4mRugoZTPPbSGWEiNKXyWLVTXypzvh0F7wDaLc849qCXU6PNbqH/r8d3dkFbJo\n6qrZlYlBPzjGNvpasUOZ3Ygz3VmulDjU5G+BjHvrv++++zh48CCBQICPf/zjXH/99STT9cRXX301\nGzZsYPfu3dx+++2YzWZuvfVWAHbs2MGhQ4cIBAI888wzAHziE5+gsbGx4MURCqhDpk8dh3wNC2c5\nokU1+syb2Gi4QtidbsDaXKsm8+eXqh6RQJ1WVShOi4EH37t4RP2+hsdmZHdHCCEEfz42SCwlOLd+\n+sTjPDYjr3WoRn5jjYP9XWF2tQe5avHom2R/LMVgNJU34aehGf0PrKvgn59q5Y3T4ZyhLhotA7F0\ntU3aKNTNh+e3qRU8A31FK1zGkqrxunxB7hBzZ1oqIhgvfmrZXOJYX5QFHgtGWaI3PHU3sK5gAkWQ\nM4NhOA1pddZWX5yqYTdVacU65Nu+rCar7Q6UL96i1t+nO3Lb/HEsBgmH2cBrHVnXl8OldiL7BhCx\nqLozCPqRSpaPuV7t3MvsJkJxBUVA2F6Ko3cKE7l33HHHmH+XJImbb755xO+3bNnCli1bCl5IXkJB\ntYTM7pjYaLE3OeJkkzoQwzbxdvnxeK0jiMdmZEHaQ692mDJDo4d7lePhGKMSx2szEkkq/OCVLrYf\n97Gq0pYZrD4deG1Gkmk7WG43srnOwVPNPuIpBXOeZB5kVXm4Rvf0N9U6CMYU1lfbqXaYOHA6zLuW\nj0xS7k3fTAejKfyxFM7aeerkrxNHVWenAE343R1BJEliQ00J7f44gpEVKI6M0Z/c0JjZpDMQ53h/\nlBvXV9AZiHOygB1ZoXSlq3JqxjD69elSzlZfLCO7nI209pyhf196DeLP/41ID3tv88Wpd5uZ57Zk\nPnNIy1S43OAbQLnrDqRNFxTk6feFhzx9rTosaPPgCBde4Ti3O3KDfqQSpypONJEpM292WpqQpqC0\nbjR6Qgl2tgU5NyvMYpCljGEpxtMfDy3evP24j8sXuLj7ynmYDNOXp8iOb3vtRs6pcxBLiRwdoeFo\nRn+s8M7mOgefubgWSZJYXGbleLppbTh7Ooe6mVt9MdXTB4SmUTNOeOcvTYPc9XQbdz3dyqvtwayy\nw9y1aTurM9noP9XsQ5bg8gUuKuwmBqOp8UXkCqQr3XcxVqjSaTFQajXQ5h9/cpa09V3q/9PGu80X\no85loaLEhC+Wym0sc5UiTh2H0x2Ifa+CohRg9NX1ltuNQx3gNpc68L5A5qzRF0KoJ1LiUFuWJzJl\n5k2MGOyHwT5oHNvotwxEJywj/B97VNnbv1qd63Vqw8ZrivT0x8KbZYTfvWL0AejT8Xpem4k11XYs\nBikT19c8QFC/aF/YdpIXTwawGmXK7YUlRBd5rfSEk/iHxdOjSYWDPREuaFDzVacGY2qDltON2P8a\nANIYidxTgzF++EoX62tKWOix8s3n29nXHUaWRnqsWsPYmWr0FSF4utnHuuoSyuwmykuMCKA/MjVx\n/c5AHLtJzhnvmY8Gt4VTo+RnspHcHuSv/Rjpb/6WSEKhJ5ykwWXGbTWgCAhkfw5ujyqfDdB+BGZI\nGwAAIABJREFUUv1/AZ6+1ShjNxkyRQPd5tKi7OOcNfpEI2qTQ4lLHRCQFd5R/vK/iKaDs7i4OUDb\nCQCkceL5//RkK4/sL77kscMf57mTft6zwjtCN/6ieS7Oq3dMqUSCZoQXea0sGKOSYqrQPH231YDJ\nIGE2yKyrKWFXW5BX24P8v98305T20vd2hjjYE2Ffd5gG9+iVO8PRJmsdH8g1Fu3+OElFcNF8J1aj\nTGvag5Q2XDA0km8MT/+xg32YDRKfvqiWz1xUSzwleKrZR63TPGJ3ZDPKyBIEY2dmTP/UYIyecJIt\njaox1AacTFUFT1cgQY3TNO5nOr/UwilfrCAJCKmyFslqoyPtONS5zZmZENkNe5LLM/K541Tv9IaT\nGaejsdRCrdPEw4k6IobCHbC5a/Qz4/8casw6Hd4RSgrx379QRYlCAZQH/xUxgeHAZzqZQdZjGAet\nvKsjMP62dDgtg6rB07zRbM6pd3DnpfVTWllTUWLCaTHwruUjvwjTgXaTyfbaz6lz0BNO8os9quHt\nSncdNw/EMMkSRlkqqoRUS+AOD/FoVUrldhMNbjOtaQ9SuuEW2HQhWKyjTpLqCSV4rsXP1UtKcVkM\n1LrMbKgpQRH5O0olSU0inqmevhZ+0c5Nc0C0jvDJH39kxVM+FngsRJMis56Cjp3+3tU6s4x+NNvT\nTyd1sxvhxq3eSWTKNU0GmdvOr+F0yshj8y4veF1z2Oinx/851Jh+xtMf7IdUCtF+ErH/NcSOJxFP\n/3EWFzpLBNOSFmNcJFpYp7uIC1WjM6A+Zyrj9mNhM8n8x3WLuWxY9cl0oXn6XtvQ+W2qVRPHrena\nak0Work/yiKvlXuvaeRD6ysoFIfZQLXDlNEl0tC0Y7w2oxo2SMfjJaMR+f99DvmbPxu1E/fRA31I\nErw7Kzl8zVLVeOTTjlHXIU+b0U+kFB4/OkAiNT07Ce3a1QxzjdOE22JgX9fkc3wpRdAdTIyZxNXQ\n6vhPDOTP0eRDq7SpsJsoteYpndU8/VUbh343zuS73nAyY/QBVlbaWW5Pcci9oOB1zWGjnzZqdmeu\nDGmfGmemuz2jySOe/0tGlvasIeRXy7ystlEfok2H6gklilLhAzXW6bYasJumXvtmNGayY1RVAiXn\nC1RmN7Eo7Z1LqF6ZIgTNAzEWei3MK7XklV4Yi0Vea0Z2WkOrtS61GqlzqdK94YT6WUmSpBYvZJFU\nBPe/3MljB/vY1jTIWxaX5oTcNtc6eN+qMi5dkN8BKDEbCExTyebzJwP8eFc3TzUXp6tVKN1BNebu\nSCekZUliXU0JeztDKEVe08PpDSdICbUibTwa3GYMUm6T2Hj0hBNYDBIlZnkUT181+tLyteBNOxNj\nOHEpRTAQSVJmy11vtU2myza2jEU2c9boi7Snj8OpxvTjMUQqhehLjyxLpRCv7VDFoXz9sG/n7C12\nNgiqjWtjhVg0Tz+WEkUnczuDiSlN1M41DLLEJ86r5u1Lc8NJ1670qiJwdiMDkSRdgQTRpDJmx+ZY\nLCu3cTqUyFRdgLqDcFrUXIIWZvJFR/98TgcTbD/u45d7ejDJEtevzg39GGSJG9dXUD9KKanTbJjQ\n7IBC0HodtjVNT4i1K5igypEbc99QU4IvlqKlCAOcD203W4inbzLINLgtxXn6oSTlJeraS8wyRhkG\nszx9qXYeGAxIy1ar1VsGg+rgjsJgNIkich0Vbf39lkJFGOaw0c9IMJQ4h96IaGTI0wfw9ata3WWV\nKH/5/YwvcTYRgfFrerMNfSFKi9l0BuIZLfE3K1sXlY4ov7x4votbNlfhsRoZjCYzXvqiPA1WhaDN\nXM0ORwxEknjTnl9pluLnaGif42WNLv7+wppxpSKGM10x/ZQi2NMZosQk09Qfpbm/cINYKN3BxAhP\nfH2NGobb01mkVPswDqcltWtdhTk3CzyWjNJnIfSGE5mckSRJuC3GHE9fqpuP/K+/QZq3CGnzxUgb\nLhjTidPCjlXD3o/q0tF3+/mYu0Y/K5E7ZPTDanWDvUTtZgNoXIx01Xug6SDi2FlU0RPyjxv/yzH6\nRcT1Y0mFvnCyIA/ozYrHZmAgkqR5IIpRHj1ePh6NHgsui4HXu4YMVH8kiSctj6DlFrI9wERK4evP\ntmUkHLQw3duXebho3tg3+nw4LDLBSQyCH42jvRFCcYWbNlRikiWeOuHL/O1nr3Xz0qnCxeLyIYTg\ndCgxogvWazMy323hwOmJx/Xj6VzEptqSnPLdsVjgsTIQSRb8XeoLJzPVRqBKYgyXw5As6nUlX3gF\n8v/7xzGP91JrAKtRYlVl7m6g2ltc5/rcNvoWm6rxrQ1ojoQRfaehshaqVK1+ad4ipIuvAocL5fcP\nI+KT2/KdMQQDRXn63UV4+tpjz2ajX2o1MhBNcXJQba6ZaKOYLEmsqbKzryucyav0R5KZwTEeqzZL\nYOiz6ggkeKUtyEutqtHUJnHlG05TCA6zgVBCmXQMfDi70w1mF813sqLClmlsG4wm+f3hAZ5p8Y1z\nhLEZiKaIp8QIzxZgXqm5oGap0Xj2hJ/BaIp3ryg8Fn5OnQOzQeLeHR0kxyndTKbj7+UlQzeU0vTu\ncSKkFMFLrQE21TqwGHPNdrWnuM71uWv0gwHVywewpbcv0TD09SCVVULtPHXgQN18JIsF6V03wOF9\nKHffkRk7NldQnv2zOuNyKgn6M11/o+GPpSgxyZSY5fHnpGbRmWlNf3OHd8bCYzPiiyY5ORhj3hha\nO4WwrrqEvkiSlsEYihAMRpIZY+9MJ5SzjYH2b60ZSGvoGa+BaDQcZrUxKFzgoJhCae6PMs9twWE2\nsKrSTstAjGA8lQllaTHzidKdvg6r8gysqXWZOR1MTLhqaFvTII2lFtZWFS5hUusyc9v5NRzsifB/\nR/LLMGv0h5MIyPH03dbc8E4xHO6J4Ium8pZQuywGSpJvguYs4R8cmiikefrhkBreKatEvuKdSNfd\nhGRSv5Dy5W9H/vjnoasdcWhujVcUL25H/M+vEPt2Tc3xhFB3Qo6xwzuBaAqnxUBliamo8E5XILdM\n7myk1GpEEWqJ3LwJhnY0NtWVYDVKfGn7KZ5r8ZMSQ2EdgyzhshhyjL6W1NVKOf3RFAZp5HD5QtEq\nX6Y6mdvmj2fUKVdW2hCoxkkLZXUG4pPaXWg18VV5nI86pxmBWnBQLL6oqhx7wbyxCyHysaXRxQKP\nhVfbx84n9GbJJWiUWg34oqmiK+mO9UX4wSudWI1SRu02G0mSqE4U7ujOWaOPf3BIczod0xenOyCZ\ngPJKpGWrka9+b+5z1mxS/9/dMYMLLYB0Tb3yy/unZhcSCRWk0+FPz6utLDEVlcg9ORjDbTFM2LN8\nM6DF3GFsrZ1CKLeb+N41C3BbjfxopzocPDuO7LEZc8I72g2gwx8nkRIE4urNe6LNcJroWmAKu3I1\nVU9NfG5ZuQ2DBAdPq7OPZQniKTFmgno82v1xZCn/aEot+do+gRDPns4QgqG+jGJZV13CoZ4IseTo\n72dvVgOeRqnVSFIRhIoonz05GONL208RTwm+fFnDqCXUVUrhSe25a/QDg0MDozVPv03Vp5C8lTxz\nwsdPdnXlPEUyW9R619NzzOgHfLBsDYSCKL96oOg7/Qi0xqxxErmBeAqXxUCdy0xHIJ65SNWhGupF\nGUkoI+KTR3ojLC2ffimEuYwnqx5/sp4+QJ3LzLUrvUST6nudLfjmHhbr1Tz9lICOQJxALDXheD5M\nj9JmZ0BV9axLG1+LUWZxmZUnm32cDiUzstgdgfiEr/d93WEWe615VU81KeSOCRj91zpCuK2GCVdk\nrau2k1QEB3tGD6lkPP2S7M9Z/RwKjeuHEym+/mwbNqPMt94yn9VjhKKqpTM8vCOUFPh94Ewb/XRM\nX5w8DkCsrIqfv3aaJ44NjtTCqKpFzKKnP/wCF4kERCNIy9civfv9sHsH4tUXijumf1Dd5Wikdwvj\n6XT4oylcVgNrquwkFTLeyRf/copP/t8JgrEUt/+xOSM7ABCMpWjzx1lWXlwZ2JsNrSzSbJDyJhIn\nwoXznFjSCeEcT99qyNFkyTYKpwZjqvTyZIx++rlTGd7RygezpR8ume9CAKsrbbx3hSoY90pbkI/9\n7jivtBVeyZNSBMF4imN9kUx55nBKzKry5Xie/vCYv1ZmuqGmZMLNgCsr7RhlidfHKBntDSexm+Qc\nz1xr0BqrJyObJ4/76Aom+IeL64Ymbo3CFnrG/Hs2c9LoEwyAUIa0KcxWkCToOAkGA9v8dlWmVDBi\nio5UWTPp8I44eRzR3YEI+Ejd9fcoz/65sOf196J85kOIva9knUvaK3e6kd7yXpi3CPHogzlVRqLj\nVOaGlg/lR99AuesOhKbEF8rqYRgDfyyFy2JkRYUdgwT7u8Pc/3Inh3sj+GMpvvVCO6dDSfZkTfQ5\n2qd6DGe90U9/Qetc5ilT/LSbDFw034ks5Xr6HptaKaQ5DL5oknqXGVlS4/qT9/TVr3lgCo1+uz+O\nRG6F1zuXe/mP65bwtavms6TMilGGx48O0BdJct+OzoK88p5Qgvf/11F+tLMLRcD66tFDMNoOdjSe\nOeHjg48e40jvkBfcnt45rRvjuONhNcosr7Dx3En/qOfkzzO6UvvZFxvf0xdC8Kejgywrt7GqgGTz\nfFvhIaO5afTTAmpaeEeSZTXEoyhQUc3/HfNlLuQRujKVtRAOIoITi52LRBzlvn9C+dn3EHtehtYT\niF89gPLM47mP27eL1N13DHUOA+KJxyDgQzQfGXpgQC1bk5wuJNmAfP1Hob8Xse1/1OeEQyj3/hPK\nD+5GKCM/OHHqODQdgngc5YdfQ3S1D53bKDH9RDqWGksJnBYDNpPM4jIbTzQN8vzJADesLaex1MLr\n6SqLNn88E+450htBlmBx2dkd3rGZZGxGeUpCO9nctKGSL15an1N2NzzWOxhNUVFiotph5mTa03dZ\nJj7fVtslTCa+PpxWf4wqh2lE+aCGQZaocphJKrCywoYi4HeH+vI+NptDPRGiScELJ9Wa9KVjOB+1\nTvOYN5I9HSGiScE3nm3LdERr4neLJxja0fjQ+griKcE/PNGSt0s3klCwDXtvMvr3BfRM7OsO0xGI\n87alBXba2guv1Z/TRj9TvQOZEE+4aj7dwQSXzFcN3nDVO6mqTv3HBL19set5dadx4ijimT9BWSU0\nLkG88Jfcx+3eAaeaEX/6L/XnwX7Ec0+of+ztHnpgMF2r7FCFxKRla2DThYg//Brl5WcQv/2pKiI3\n2A/Nh0eu5+k/gdmijmQLBVD+5XbEjqfSx8xv9H+59zS3/kHV6dYutDVVdgKxFNUOE9et9PK2tPzA\nhfPU3cLhdHzySK9ahjeTmjtzlU9dWMP1q4sfUD4WpVbjiOlLpcNivdqA8wUeC839UQKTDO+YDTKN\npRYOjRGDLpZ2fzwTzx8NbTDJe1Z4qXeZC5JDPpFuhnOaZdZVl4zZH9HgtuCLpTJaRsM50hdhgcfC\nQDTFsydUR+n4QBSzQRp37eOxrNzGt98yH4tB5l+ebhtREh1JKthMoxj9AsI7L5z0U2KSM9/P8ZCW\nrCpw5XPU6GeGBWfrTaeTuV2VCwFYXaWGLEZ4+ummrYnE9UXAh3jqj0Nyxa0nkNadq2pjtLfkiLqJ\n5rTY21N/QPR0IZ76P1X/v6oOkWX0M9U6ziH1SPnDt6thnp99D7HjSaQtbwGjSdUSyl5P32nEK88i\nnXcp0ppNyP/yQ1Wj48j+MXU6mvqimZps7ULbnK5U+MjGSkwGmSsWurn13Go+eV41RlniYE+EQz1h\nXu8KsaZ6+sYvnkmc1+DMjMqbTrRQz0A0iRCCwWiKUquRxekhLCkx8cYsjdVVdg71RKZk4pQQoiCj\nv6TMRrndyMZah9rhXEACs3kgxjy3hXvftoDbzq8Z87ErK1VHMN+0M180SWcgwZb5LjX2nw4DHe+L\nssBjnZKQXY3TzFeuaCCWVLj/5c6cfF4+T99kUHePhXj6R3qjLCu3jTq6cziSVrlYAHPS6Gue/oDZ\nybeeb1erDtIGrrNU9eTrXWYqHaaM5nmG8ipVoqG9paiXVJ55HOXTN8LJJqRr3geL1AHF0rpzoWEh\nJJOZ4ewiEoauNtVYyzLKo79QdwLrzlFvED1ZVUV5QjGS1Y78qX9B+sjfI3/uHqQP3gqrNiBe25ET\n4lF+81OQJKS3/7X6vFIv8odvU5vSSvLXGAshaPPFMp6hJs60otLOL69dzPnp5g6TQeItS0opMRtY\nWmZlxyk/336+g4oSEzesGXtUn87UoiWNByIpIkmFeEqo1SVZIbbJls+urrQTTwma+ifv7fdHksRT\nYtyO7b9aXcYD71yIySCp3ajjhJeEEJzoj7LQa83MVxiLhR4rJSaZ/d0jE6paHH9ZhY06l5k2Xzyj\nmLrYO3U38vmlFj64roLXu8LsaB1KVufz9EGN6/vGMfqRhEKrLzZtFXQTDxROJ/5BMJrYO6jw4qkA\nWxpdnJOWEO60qNvtGqeZKod5hKcvGU2wbDVi2/+gxGNIN9yiDiEeB7Fvl9r09cG/U/WtHU5EKgVL\nVyH1dCEAcaoZqb4RWo6BEEgbLwSHG/EntdtWvuxtiFPNEPQjomFVPiLgU29CJblbesnuQLrwyqGf\nN12EeH2nGlby9asdvKeaka69CalsSMNdql+A9N4bhyYsDX/rYikCcYWPbqxkdZWdhZ6hC3w0oa7z\nGhz8ck8PNU4zn76wlhKzHtqZSbRKnv5IIlPZUWo1sihL2XOynv6qtFe8/biP5v4Yb11SOmFvd0jj\nfuyKEoMsZV7DYzNmZsSO9rr9kSS+WIoFnsKMskGWWFlpZ38eT/9IbxSDpMbu610WdrQG6AjEiSaV\nCZdqjsZbl5Tyl+OD/GpvT0YbKZLIb/SdFsO4nn5TfwRFwNKy6SmmmLtG3+3JxOtbfTHOtdrVDjzJ\nRpktgcUoU+0w8eKpkUkU+bYvqx2wf/lfiEXhw3+vJoNHQSgKHD+MtPECpNXqNkk65xI45xL171W1\nYDbDsQMoh/chtCHEC5YgLVyGeP4JdSeyYp3aNQxqXL9+gVpe6XCO+fqg7iiEwYjY9bxa0mm2IL3r\n/UhXvWvk+b31ulGP05ZVSlfoxf2eFWW8a7l3RvXsdYYoMclYjTK9oWQmrl9qNeCwqENYuoKJSXv6\nLquR+aUWth/3AT7qXGbW15TQE0rw8L4ePn5O9ahJ2eFkOmWL6NjWOpwDsdSozoemVV/MuMw1VXZ2\ntQfTipZDN6GjfREaPVYsRpk6l5lALJWpUptqo2+QJa5aVMq/vdqdVqc15w3vALgthlFzEBpHelWb\ntmSaKujGNfoPPPAAu3fvxu12893vfnfE34UQPPjgg+zZsweLxcKtt97KwoVq3P2ZZ57hscceA+Da\na6/lsssuK2hRwqdKMGhyAK2+OJR6wVtBRwRq0rHEqhITgViKUDyV451KZgvS9R8jZbGz66W9bNz9\nEqbNF418nXAQ8dADSOdtUYewL16Rdz2SbIC6xtxkbnUdUjpjLn/6blUXW5ahogoB0KMafREcXwIZ\nQLKXwMr1avI4lUL+xBeR1p9X0PuVTatf/eKMpq0+GrrBnz0kSaLcbqQ3nMxos2glo4u8VrqCiUl7\n+gAfWFdOU1+U/znUz672IOtrSni+xc9TzX62LiwdtTRQCJETSuwOquWa+TplR0PrcB6IJkc1+jvb\nVGeqUE8fhqSrD3SHuTRr6lpPKJGp0NFyD384MoDbYpiwYupYbEznzHZ3hHjrEhOxlMhbDOGyGjgx\nzoD1o70Rap2mKfnM8zHurf2yyy7jzjvvHPXve/bsoauri+9///vccsst/PSnPwUgGAzy6KOP8vWv\nf52vf/3rPProowSDwVGPk0Nad0cT/mr1xZDedQPyP36DzkCc2rQWh6bJ8cSxwbwt0a9vfjtfX/MR\nHtuXPxQiXnoa8eoLKD+/DwBpUX6jDyA1qDcy6bJrkN56HdJb3zf0t/pGpJoG9YeKavXYWjI36MtJ\n4o6FtOkiNRlc6oU1mwt6znDa/HEsBimnE1Bn7qMa/UQm7q3VdK+osGGUpSkZQn9evZMPrKtgXbXq\nHQshOJLuyxht5uyxvgg3PHIsMyQeVG2mcruxKOXRITXR/F7uE8cGeaJpkLctLS2qckzT/smu4hNC\n0BdOZsJmmtHvDiY4v8E5ZX0X2dQ4zVQ7TOzpDBJN26J84R2XxUggNrr+TkoRHO6JjFmqOlnGNfor\nV67E4Ri9BvTVV19ly5YtSJLE0qVLCYVCDAwMsHfvXtauXYvD4cDhcLB27Vr27i1QCK23C8lVmvkg\n2/xxFGsJIWcZ/lgqk0BaW1XCAo+FX+7t4fsvd444zOvd6oX6mHEh/b7cZI8QQvXcZVkdxehwQVXt\nqEuSNl8EqzchXfdh5OtuQr7oyvwPtKf1/7VkbgHDTjKvsf48sFiRLnkLkmFiX/I2n1pVoXvuZxZl\ndhN94SSnQwmMsirNAHDNUg//+vbGKS2h3VznoDuYoNUXz4QSRlPEfL0rTCSp8Is9pzOGqiuYoKpI\n2e3SzNyA/PHs3+7vZVWljZs3VRV1XLNBptRqyNGWCiXUZLjWxVpZYsKYNvSFlkBOhA01JezvDmfm\nH+Q3+gbiKZGR4xjO/u4wvliK8+unb52Tdgf7+/spLx+q9igrK6O/v5/+/n7KyoZqnL1eL/39+eVI\nt2/fzvbt2wG45557kIxGpHXn438jxTyPjVMDEZIWJxGhfrDL68opLy+jHHjoxkp+suMkD73axofP\nt7CieujNOtDbSoNdoiNo5JY/nqLKZeOyxeV80NiK1HKEUFsLjg99gvDvf41pxVpKK8YYen3Jlep/\nBdBXXQfHD1FqkOgLBbBWVOEqL6Aipryc1I8fRXa6J2z0O4InWFvryvlMdOY+88pDPNXsoy2oMN9j\np7py6Fqsrpza17ra4uRHO7t5vDmU8bz74xL7B2BPu48PbKqnyqmGQNpDvYBqjI4FDVy4wEtP+DgX\nNHqLusbsrhTQTFy2jHheKJakL5LkfRvqqKosfPC8Rq27jcG4lDmuv0918OZXejK/a/C00heKc9nK\nBowFlkEWy0VL4PFjg7RFVbNa6XGPONe68iTQg7HERblrZG7h5d39lJgNXL12fsE5lmKZEzGArVu3\nsnXr1szP8r3/qY6pe6OFTdWq0X/9RGdGGMyaitDb25t5/NsW2vjf/QY+9/sDVDlMfHhjJTUOE8d6\nQnxgTRmL/vt+9lWvpqXqPH71Whvew49wWddrhO2l/My4ivd96pt4XLacY04G5cp3IX7xfXo/+TcQ\nCRM1mYkXc+yBgQm9biCWojsQo9rKlJ2LzsxglxIIYG+7j/MbnNP6+cnABQ0O/nRIDXt6rAZaegM8\n2BfkeH+UP7zRxT9cXMu59U4Od/k4p66EjkCCb/zlKPdcPY++cAK3KVX0Gq1GmbY+H729ucbuWDrE\n5DEkJ3TeHovEiYFQ5rnNaU0cU3LITrxnmRtFwODA2Dr4k8GcUs/jWKfaeZyMhujtzd1xSwn1MS2d\nPZjiuSGcWFLhmWO9XDjPSWCwn2LnjtXWjh6pyGbStxKv15vzQfX19eH1evF6vfT1DbVd9/f34/UW\nPqVGi+efk1bra/XF6cvIlebeq+wmA3+7uYoqh4nuYIJ/fqqVX+5VBYjW1TrY8LYr+dDuX/Gl449g\nVhI0exci3/VDfvNXd/HH5iC7QpZxB5IUg3zepchf+VdYvBKEUKUhZoCmfi3rf3ZLKJyJaNd0PCWY\nP0kp50L44PoKZAlMssS59U7a/HFODES5apGb+aUWvvFcO8+3+OkIJFhSZuMfLq4lEEvxT0+2AhOb\nteCxGfLW6muiaRPtkq0oMdETSuZMJgMy08kALlvg5oqFheXWJoo7nXjVClDyV++oa8rXldvUHyWS\nVDi/objxh8UyaaO/efNmnnvuOYQQHD16FLvdjsfjYf369bz++usEg0GCwSCvv/4669evL/i4mtFf\n6LHisRroCKhG3yRLecvXtjS6+MbV8/nWW+fjtRnZftyH22pgsdeKfN6lSFvfhbzreRpDnZxYsJGT\n1gr+lC73bPdP/YhFqboew21fRv7uL9XqoBngWLohZapL0nSmn2wVxfkz0AVc77Lw12vK2brITYPb\nTDwlUARc0ujirisbqHeZ+cErap5socfKAo+VOy6sQQAS5PR/FIonPYJyOJpu/kQntVWUGEkoItPj\noOnsFDr7dqrQ8jBaLnK05iwgb4OWVr9fPo6i5mQZ91257777OHjwIIFAgI9//ONcf/31JNNyBFdf\nfTUbNmxg9+7d3H777ZjNZm699VYAHA4H1113HV/4whcAeN/73jdmQjgbfyxFZyBBqVUVC6t0qKPR\n4jaB124cc5hEud3E/e9YgC+aymkOkf/6ZsS7P8Ci1wd59mSAxw72YzPKOCzjy7NOBilbSmKaOdYf\npc5lzuin65w5ZO9eZ8LTB/ibdOf1q+1qVZ1BUjVlrEaZ96+t4J7n2wFYmO5gvXi+i4vmOYkklQkl\nlkttxswIyGza/XEqS0yYJhhrr0iXjp4OJSi1GemPJHGY5WmLiY+G1ShhNkgZlYDRmrNgaO5xNtrw\n+ukeXjSu0b/jjjvG/LskSdx88815/3bFFVdwxRVXFL2oloEobf5YRvekqsTEkb4IihCUFXD3liUp\nR7o2s1arjUVlMR5v8vHCST+XL3QTTih5VfLONIQQHOuNTEoyVmf2sKcbtCRUz3Um0arhFpdZsaYN\n5XkNDhpLLQxGkzkesyRJE64k8lgN7AknR9T9dwTG1/EZC61foCeUYGm5LadccyaRJAm3xUBPOgyd\nL7xTYpIxSPk19bWqn+l22uak9s6JgRitvnhmIHWlw0RPKEFPOJnRkpkoC9Ohj5RQhz7UOVUph6kQ\noppN+iJJBqIpPZ5/hiJJEhUlRuaVWiY8FnGiVJaYsBol1lYNOQyyJPH5LXV8YUv9lK1K7Wb5AAAX\njElEQVSnzmUhklQyowQBlLR4W+0kjH62pw9qTH82jD6onc8a+Tx9SZKocaqS2cMJxFMYZXXHMJ3M\nieqd4ezuCBJOKBkt8yqHCUWozRX5psEXwzy3GaOsTt5ZU2VnIJJEEdAVjOO1GbGZ5DOyxv1Yunlm\nyTTpdehMPx/bVIW1iIanqcJkkLj3mgUjHKoap5maKSwX1zptWwZiGUPdF1bF2+qKrPvPxmE2YDfJ\nGXnj/kiSBvfs7Hg1mWxZUqeu5WN9TQnbmtSG0uwQVDCewmGe+CzkQpmTRn9fWkBJa5fObvee7B3c\nZJC5tNFNvVudiKRtK//1pU6O9UUxyRJ3XdnAysozS174WG8Eg1RcC7vO3GLDKKMBZ4LJeNqF0pi+\nNk8MRjNVec3p0Opk8xgVJSZOhxKkFHWA0Kx5+ul4vM0oj2q8N9aU8H9HBjjYE8n5zAMxZUbycXMy\nvKONvdXCO9kzSicb3gG4/YIarl2pNo5pRv9YX5T11XYUITKJrTOJY/1RGj2WgvW3dXRmGrtJFZBr\nGRgKbRxNq2FOtuKswa2GTAai6s59KuzERNAqeKx5Qjsaq6vsmGSJPR25diYYn9xYzEKZsxbCbTFk\n4mPldhOaXMZUf5jagOUSs8ynLqplodeaM1PzTEARgqa+qB7a0ZnzNHosGTVNUHXvNTXMybDQY+V0\nKMnedGPWbO143Vme/mhYjDKrKm282hHK0eAJxlOZIfbTyZw1+g1Z2z2TQcps18psU1/D+pGNlXz2\nolpKrUaWlds42hfNdP+eCXQE4oQTCkvO8rm2OnOfBaVWOtO69ilFcKwvwrIpKD7QjPyfjw1ilGev\nV8WVjunnS+Jmc0mji3Z/nNc6Qjy4+zSvtQcJxFIzEt6ZkzF9owzz3bkxxiqHid5wMm8p5mS5LEuS\ndXm5jf87MsCJgTPHcz7Wqydxdc4MGj0WBHDX062sqLATTQqWTYGipFaVd6wvypIy66yFOTVJ7PGM\n/qWNbn69r5dvv9BBNKlwOpRQZyGbp3/dc9LT//JlDVy3KncgdZ3LXLSc60RYXqFegGdSiOdQTwSr\nUaJ+BpJxOjqTYU2VnfPqHQxGUzx6QJVpmQqjX2o1ZqIBU3G8ieIqILwDavTivSu9RJMKsgSnBmPE\nUmJGwjtz0tNfn6eK4YPrKnjX8sK1eyZKRYmJMpuRIz1R3rFs2l9u0gxGkjx9wseF06QTrqMzlZSY\nDdx5aT0pRfCLPac5ORgbd+xioSz0WOiPJGfV6LsLDO8AXLPEQ0WJiT0dIR4/ps4Fd56t4Z18uK3G\nTGZ8umn0WDITqOY6jx3sI6kIrteHmeucQRhkiY8VqZ0/Hgu9Vl7tCLF0FnNbmo0az9MH9T04r96Z\nEZKE6ZdggDPI6M8ktU4zb3SHR7SLzzVSiuCJpkG1s1gP7eic5bx9mYc6l5nqSTR6TRZrenZ3Md/H\n2qz1nrWJ3NmmzmUmlhL0pZs8fvpqN7UuM+9YNv3hpWJoD8SJJsWsNvXo6MwVSq3GnKKM2eKH71xI\nManHbHXRmfD052Qid7bRuhM7/HF+va+XPx4d5A+HJzbYZDo5kdbP17twdXTmDkZZKipCUG4fGufo\nOFurd2YbbWv2cmuAR97ow2sz0hVM0JvW6Q4nUqTmQB1/y2AMoyxl1Eh1dHTOPAyylElm657+LOG1\nGbEYJLY1+ZCAT55XDcALJ/383e+bueGRY9z30shB7KMhhMjpvJsqmgdiaQG5uZt30NHRGZ9alxmD\nVFgCeLLoRj8PsiRR6zKTUASrKm2srymhxCTz0N4euoNxFnosHDgdLvh4P3ili8/8uSUz0WeqaBmI\nssCjd+Hq6JzprK8uYWWlfUYKR3SjPwpaRv3CeS4MssTKShtJBa5eXMqWRhd94ST+6MjpN/k42hvh\neH+Mf/jzSfaktUEmy0AkyWA0pcfzdXTeBLx9mYevbp03I6+lG/1RmF9qQZbIDCk+t96J0yxz/Zry\njHd9Is8ghHz0hZOsr7ZjNcn881OtvNxa7Jz7kWjTvhp1o6+jo1MEesnmKLxzuYdNtY7MwOqrFrm5\nfIEbk0HKlGO1DMTGHU8YTqQIJRTWVZfwjuUePvRoE/u6Qpw/yWEwTenKHX0Iuo6OTjHoRn8U7CYD\ni8uGMumSJKGNBnVbjZTZjJkBEGOhdduV2Y2YDTL1bjNtUzCIvalPHYI+0XmlOjo6Zyd6eGeCDNcF\nV4Rg+/FB/v6PJ9jfPRS31+aBlqenf9W5zLT5Jm/0j/VFWaJ7+To6OkVSkKe/d+9eHnzwQRRF4cor\nr+Q973lPzt97enr40Y9+hN/vx+FwcNttt1FWpqpk/upXv2L37t0IIVizZg0f+chH5rS0QaEs8FjZ\n29lHZyBOjdPM080+7n+5C4Btx3ysSQ+Z1ip2ytPDXxpcFp454SecSE3YS+8LJ+iPJFms6+fr6OgU\nybievqIo/OxnP+POO+/k3nvv5cUXX6StrS3nMQ899BBbtmzhO9/5Du973/t4+OGHAThy5AhHjhzh\nO9/5Dt/97nc5fvw4Bw8enJ4zmWG2LnJjNxv4pydbGYwkOdwbwWmWuWKhm9c6gpkhLL0h1dP3poe/\n1KfnBEzG29fi+brR19HRKZZxjX5TUxPV1dVUVVVhNBq58MIL2bVrV85j2traWL16NQCrVq3i1Vdf\nBdQ4eDweJ5lMkkgkSKVSuN2zr40xFdQ4zXzl8npOhxJsb/ZxcjDG/FIL59U7CCUUDqbr+HvDCUqt\nhswcAE3zfjJx/aa+KLKkjojT0dHRKYZxjX5/f38mVANQVlZGf39/zmPmz5/Pzp07Adi5cyeRSIRA\nIMDSpUtZtWoVt9xyC7fccgvr1q2jvr5+ik9h9lhSZqPWaeZwT4STg3Hml1pYX1OCSZZ45oQfIQS9\n4STl9iFBpWqn2nnX5pu4dPPhngjzSy2Tniuqo6Nz9jEl1Ts33ngjP//5z3nmmWdYsWIFXq8XWZbp\n6uqivb2dH//4xwDcfffdHDp0iBUrVuQ8f/v27Wzfvh2Ae+65h/LyM0cbfl19P08e7SGeEqyqL6e+\nupKrlg3yp0On6YkKekNJFleU5JxTg+cU7SFBWVlZ0fmNWFLhcO8Rrl1bc0a9Tzo6OnODcY2+1+ul\nr68v83NfXx9er3fEYz772c8CEI1GeeWVVygpKeHJJ59kyZIlWK1qGGLDhg0cPXp0hNHfunUrW7du\nzfzc29s78TOaYRqdEvGUGr8vMybo7e3lbzd4WOiSeWBnF4qADdW2nHNaW2nl94cH+Oc/vsEnz6tG\nESBJqvyDxuGeCN98vp3vXdOYMxd4X1eIeEqwxC2fUe+Tjo7O9FJbW1vQ48aNDyxatIjOzk5Onz5N\nMplkx44dbN68Oecxfr8fRVEA+N3vfsfll18OQHl5OYcOHSKVSpFMJjl48CB1dXXFnsucZnnWaLZ5\npWq8XpYkrlpcyo3rK4Chck2ND2+o5J3LPWw/7uN4f4wHdnbx1Wdyk+OvdQTpjyQ52BOmJ5TIdODu\n7w4jS7CyUh+CrqOjUzzjevoGg4GPfvSjfO1rX0NRFC6//HIaGhr47W9/y6JFi9i8eTMHDx7k4Ycf\nRpIkVqxYwcc+9jEAzj//fN54443MLmD9+vUjbhhnOg1uCzajjNMijyjBfO8KL6VWIxtrc7t2DbLE\nO5d5+MPhAY70RtjVFoRhUZ7mdIVOU1+UbU0+9nWF+Ltzq3mtI8iSMqvelKWjozMhJDEdmr+TpKOj\nY7aXUBQ/eLkTi1HmbzcXPvNTCMGHH2uiymHmSG8EgEf+emkmOfvhx5oYiCRZWWGjqT+KEJBIl4He\nsLacv9Fn4uro6GRRaHhHl2GYAj55fk3Rz5EkiaXlNna2BTO/6w0nqXOZ6Y8kGYgkMRskDvaoN4TP\nXVKLJEl4bUYW6524Ojo6E0Sv+ZtFlg5rruoJqd27WmjnwnmqKJtBgg01Di5ocLKs3IZBH5qio6Mz\nQXSjP4ssTSeB56W7dDWj39QfRQKuWlQKwLJyGzaT/lHp6OhMHt2SzCJLyqxYjap0gyxBT1qn58RA\nlFqXmWXlNlwWw6RlmHV0dHQ09Jj+LGI3GfjJuxbitBj4w5GBjKff5otT7zJjMkj827sXYTHq4Rwd\nHZ2pQff0Z5lSmxGDLFFhN9ETSpJSBF3BOHVpjR6bSc5p2tLR0dGZDLrRnyNUlpjoCSXoDiZIKmSM\nvo6Ojs5Uohv9OUJ5iZHecIJWvyrEVu/SZ9/q6OhMPbrRnyNUlphIKrC/S5Vk1j19HR2d6UA3+nOE\nVZV2AP5yfBCXxYDTosss6OjoTD260Z8jzCu1sLTMSjQpdC9fR0dn2tCN/hziqsVqM5Zu9HV0dKYL\n3ejPIS6e78RrM7KyQpdN1tHRmR50lc05hhCi6GlaOjo6OlM2REVnZtENvo6OznSiG30dHR2dswjd\n6Ovo6OicRehGX0dHR+csQjf6Ojo6OmcRutHX0dHROYvQjb6Ojo7OWYRu9HV0dHTOIuZkc5aOjo6O\nzvQw5zz9z3/+87O9hAnzk5/8ZLaXMOuc7e+Bfv76+c8WhdrOOWf0z2Q2bdo020uYdc7290A/f/38\n5zq60Z9CNm/ePNtLmHXO9vdAP3/9/Oc6c87ob926dbaXoKOjo3PGUajt1BO5Ojo6OmcRxtlewJnK\nAw88wO7du3G73Xz3u98F4JFHHuHJJ5/E5XIBcMMNN7Bx48bZXOa0ke/8AR5//HGeeOIJZFlm48aN\nfPCDH5zFVU4v+d6De++9NyMNHg6HsdvtfPvb357NZU4b+c6/paWFf//3fycej2MwGLj55ptZvHjx\nLK90ehjr/KPRKBUVFdx+++3Y7fZZXukwhM6EOHDggDh+/Lj49Kc/nfndb3/7W/G///u/s7iqmSPf\n+e/fv1/cddddIh6PCyGEGBwcnK3lzQj53oNsfvnLX4r/+q//muFVzRz5zv/uu+8Wu3fvFkII8dpr\nr4mvfOUrs7S66Sff+X/+858XBw4cEEII8eSTT4pf//rXs7W8UZlzMf0zhZUrV+JwOGZ7GbNGvvPf\ntm0b7373uzGZTAC43e7ZWNqMMdY1IITgpZde4qKLLprhVc0c+c5fkiQikQig7nQ8Hs9sLG1GyHf+\nHR0drFixAoC1a9fyyiuvzMbSxkQP70wxTzzxBM899xwLFy7kQx/60Fl1Y+js7OTw4cP85je/wWQy\nceONN75pt/bjcejQIdxuNzU1NbO9lBnlpptu4mtf+xoPPfQQiqLw1a9+dbaXNKM0NDSwa9cuzj33\nXF5++WX6+vpme0kj0D39KeTqq6/m/vvv51vf+hYej4f/+I//mO0lzSiKohAMBvna177GjTfeyL33\n3os4S+sEXnzxxTe1lz8a27Zt46abbuJHP/oRN910Ez/+8Y9ne0kzyt/93d+xbds2Pve5zxGJRDAa\n555frRv9KaS0tBRZlpFlmSuvvJLjx4/P9pJmFK/Xy7nnnoskSSxevBhZlgkEArO9rBknlUqxc+dO\nLrzwwtleyozz7LPPct555wFwwQUX0NTUNMsrmlnq6ur40pe+xDe/+U0uuugiqqqqZntJI9CN/hQy\nMDCQ+ffOnTtpaGiYxdXMPOeccw4HDhwA1NhmMpnE6XTO8qpmnv3791NbW0tZWdlsL2XG8Xq9HDx4\nEIA33niD6urqWV7RzOLz+QB11/vYY49x1VVXzfKKRqLX6U+Q++67j4MHDxIIBHC73Vx//fUcOHCA\nlpYWJEmioqKCW2655U2byMp3/lu2bOGBBx7g5MmTGI1GbrzxRlavXj3bS5028r0HV1xxBT/84Q9Z\nsmQJV1999WwvcVrJd/61tbU8+OCDKIqCyWTi5ptvZuHChbO91Gkh3/lHo1GeeOIJAM4991ze//73\nI0nSLK80F93o6+jo6JxF6OEdHR0dnbMI3ejr6OjonEXoRl9HR0fnLEI3+hNg586dXH/99bS3t8/2\nUnR0dHSKQjf6E+DFF19k+fLlvPjii0U9T1GUaVqRjo6OTmHMvXaxOU40GuXw4cN85Stf4Zvf/Gam\nVPORRx7BarXS1dXFqlWruPnmm5FlmRtvvJGrrrqK/fv387GPfYzly5fP9ino6OicxeiefpHs2rWL\n9evXU1tbi9PppLm5GYCmpiY++tGPcu+999Ld3c3OnTsBiMViLF68mG9/+9u6wdf5/+3dXUjT7R/H\n8fe2slzm88gQKmsVdSY9bFlTo+ggpaKDHiBIlAo6qoNiJz0chCdhg9Bh1CTpwKCTwIMOIsgoTKQH\nqHDobBGptZnTWGzB3O4D+Q/998+/3ty1G36f19musYvrOvns4rvfrq9Ixin052n6nSoVFRU8ffoU\nALvdzrJlyzCbzWzfvh2/3w+A2WzG6XRmbL0iItOpvDMP0WiUt2/f8vHjR0wmU7pGP1ujlIULF2I2\n67tVRP4dFPrz8Pz5cyorKzl58mR67NKlS/j9fgKBAKFQiOLiYrq7u9m1a1cGVyoi8r8p9Ofh2bNn\n7N+/f8aYw+Hg4cOH2O12fD5f+ofcrVu3ZmiVIiK/prt3/gHv3r2js7MTt9ud6aWIiMxKxWYREQPR\nSV9ExEBU05/F6OgoLS0tjI+PYzKZ2L17N3v37iUajeLxeAiHw9hsNs6ePUtOTg5DQ0N4vV6CwSBH\njhxh3759wFRDEY/Hk543FApx6NAhampqMrU1ETEonfRnEYlEiEQirF69mlgshtvt5ty5czx+/Jic\nnBwOHDjA/fv3iUajHDt2jImJCcLhML29vSxZsiQd+tMlk0lOnTpFY2MjNpstA7sSESNTTX8WBQUF\n6a4/2dnZlJaWMjY2Rm9vL1VVVQBUVVXR29sLQF5eHna7HYvF8ss537x5Q0lJiQJfRDJCoT9HoVCI\nYDCI3W5nYmIi3QYxPz8/3RdzLqb/o1dE5E9T6M9BPB6nqamJuro6rFbrjPdMJtOce2AmEglevHih\naxlEJGMU+v9HIpGgqakJl8uFw+EApso4kUgEmKr75+bmzmmuV69eUVZWRn5+/m9br4jIbBT6s0il\nUrS2tlJaWkptbW16fPPmzXR1dQHQ1dXFli1b5jSfSjsikml6emcWfr+fixcvsmLFinQJ5+jRo6xd\nuxaPx8Po6OiMRzbHx8dxu93EYjFMJhOLFy/m2rVrWK1W4vE4p0+fprm5+acSkYjIn6LQFxExEJV3\nREQMRKEvImIgCn0REQNR6IuIGIhCX0TEQBT6ItO0tLRw9+7dTC9D5LdR6Iv8DZcvX+bRo0eZXobI\nvCn0RUQMRE1UxNCCwSCtra2MjIxQXl6e/ud1NBqlubmZgYEBkskk69ev58SJExQVFdHR0UFfXx8D\nAwPcvn2b6upqGhoaGBoaoq2tjffv35Obm8vhw4epqKjI8A5FZtJJXwwrkUhw9epVXC4XbW1tbNu2\njZ6eHmDq3qXq6mq8Xi9er5esrCx8Ph8wdRXHhg0bqK+v586dOzQ0NBCPx7ly5Qo7duzg1q1bnDlz\nBp/Px6dPnzK5RZGfKPTFsPr7+5mcnKSmpoYFCxbgdDpZs2YNAEuXLsXpdLJo0SKys7M5ePAgfX19\nv5zr5cuX2Gw2du7cicVioaysDIfDQXd395/ajsicqLwjhhWJRCgsLJzRD6G4uBiAHz9+0N7ezuvX\nr/n+/TsAsViMZDKJ2fzzWSkcDjMwMEBdXV16bHJyksrKyt+7CZF5UuiLYRUUFDA2NkYqlUoH/9ev\nXykpKaGzs5Ph4WEaGxvJz8/nw4cPnD9/nv/cT/jfjXOKiorYuHEjFy5c+OP7EJkPlXfEsNatW4fZ\nbObBgwckEgl6enoIBALAVLe0rKwsrFYr0WiUe/fuzfhsXl4eX758Sb/etGkTIyMjPHnyhEQiQSKR\nIBAIqKYv/zq6WlkMbXBwkBs3bvD582fKy8sBWL58OXv27OH69esMDg5SWFhIbW0tN2/epKOjA4vF\nQn9/Py0tLXz79g2Xy0V9fT3Dw8O0t7cTCARIpVKsXLmS48ePs2rVqsxuUmQahb6IiIGovCMiYiAK\nfRERA1Hoi4gYiEJfRMRAFPoiIgai0BcRMRCFvoiIgSj0RUQMRKEvImIgfwEUE9ymRhANzAAAAABJ\nRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEwCAYAAABbv6HjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEwCAYAAACwiBrgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGQn8QgjRyUjgF0KITkYCvxBCdDL/H0V3nzyv/seHAAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEyCAYAAAABVZAhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhBAuSIK7EEK4IAnuQgjhgv4ffsieATy8ZSkAAAAASUVORK5CYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# env_test = env\n", + "\n", + "# Test\n", + "for i in range(10):\n", + " model.train(False)\n", + " state = env_test.reset()\n", + " for i in range(250):\n", + " state = Variable(torch.Tensor(state).unsqueeze(0))\n", + " mu, sigma_sq, v = model(state)\n", + " eps = torch.randn(mu.size())\n", + " action = (mu + sigma_sq.sqrt() * Variable(eps))\n", + " env_action = action.data.squeeze().numpy()\n", + " state, reward, done, info = env_test.step(env_action)\n", + " if done:\n", + " break\n", + "\n", + " env_test.plot()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T08:03:17.463005Z", + "start_time": "2017-08-05T08:02:49.857Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T06:59:47.486363Z", + "start_time": "2017-08-05T06:59:47.404265Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T06:55:09.249081Z", + "start_time": "2017-08-05T06:55:09.199392Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "jupyter3", + "language": "python", + "name": "jupyter3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.5.3" + }, + "toc": { + "colors": { + "hover_highlight": "#DAA520", + "navigate_num": "#000000", + "navigate_text": "#333333", + "running_highlight": "#FF0000", + "selected_highlight": "#FFD700", + "sidebar_border": "#EEEEEE", + "wrapper_background": "#FFFFFF" + }, + "moveMenuLeft": true, + "nav_menu": { + "height": "85px", + "width": "252px" + }, + "navigate_menu": true, + "number_sections": true, + "sideBar": true, + "threshold": 4, + "toc_cell": false, + "toc_section_display": "block", + "toc_window_display": false, + "widenNotebook": false + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/pytorch ppo-not_shared.ipynb b/pytorch ppo-not_shared.ipynb new file mode 100644 index 0000000..cb99670 --- /dev/null +++ b/pytorch ppo-not_shared.ipynb @@ -0,0 +1,2199 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Pytorch is easier to debug, I like it.\n", + "\n", + "TODO:\n", + "- [x] prioritised experience replay, need to grab loss for each sample, and sample based on loss\n", + "- [ ] check it for my data, can it overfit?, does the normalisation make sense?\n", + "- [x] better metrics\n", + "- [ ] do cnn model\n", + "- [x] read papers\n", + "- [ ] check i'm prioristising by the right things, should lead to lowest loss\n", + "- [ ] test on cartpole\n", + "\n", + "Refs: \n", + "- implementations:\n", + " - PPO\n", + " - **pytorch implementation https://github.com/alexis-jacq/Pytorch-DPPO/blob/master/ppo.py**\n", + " - tensorflow implementation https://github.com/reinforceio/tensorforce/blob/master/tensorforce/models/ppo_model.py\n", + " - Prioritised memory\n", + " - Other\n", + " - http://pytorch.org/tutorials/intermediate/reinforcement_q_learning.html#training\n", + " - https://github.com/pytorch/examples/blob/master/reinforcement_learning/reinforce.py\n", + "- papers:\n", + " - DPPO https://arxiv.org/pdf/1707.02286.pdf\n", + " - PPO \n", + " - https://arxiv.org/abs/1707.06347\n", + " - https://blog.openai.com/openai-baselines-ppo/\n", + " - TRPO https://arxiv.org/abs/1502.05477" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:11.868333Z", + "start_time": "2017-08-05T14:42:11.056832Z" + } + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:__main__:__main__ logger started.\n" + ] + } + ], + "source": [ + "# plotting\n", + "%matplotlib inline\n", + "from matplotlib import pyplot as plt\n", + "import seaborn as sns\n", + "plt.style.use('ggplot')\n", + "\n", + "# numeric\n", + "import numpy as np\n", + "from numpy import random\n", + "import pandas as pd\n", + "\n", + "# utils\n", + "from tqdm import tqdm_notebook as tqdm\n", + "from collections import Counter\n", + "import tempfile\n", + "import logging\n", + "import time\n", + "import datetime\n", + "import random\n", + "\n", + "from collections import OrderedDict\n", + "from IPython.display import display\n", + "from pprint import pprint\n", + "\n", + "# logging\n", + "logger = log = logging.getLogger(__name__)\n", + "log.setLevel(logging.INFO)\n", + "logging.basicConfig()\n", + "log.info('%s logger started.', __name__)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:12.275405Z", + "start_time": "2017-08-05T14:42:11.871092Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "import argparse\n", + "import os\n", + "import sys\n", + "import gym\n", + "from gym import wrappers\n", + "import random\n", + "import numpy as np\n", + "\n", + "import torch\n", + "import torch.optim as optim\n", + "import torch.multiprocessing as mp\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "from torch.autograd import Variable" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:12.325360Z", + "start_time": "2017-08-05T14:42:12.277635Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "import os\n", + "os.sys.path.append(os.path.abspath('.'))\n", + "%reload_ext autoreload\n", + "%autoreload 2" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:12.406770Z", + "start_time": "2017-08-05T14:42:12.326902Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "'outputs/agent_portfolio-ddpo/2017-07-21_seperate_weights.pickle'" + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "class Params():\n", + " def __init__(self):\n", + " # env\n", + " self.window_length = 50\n", + " # Model\n", + " self.batch_size = 250\n", + " self.lr = 3e-4\n", + " self.gamma = 0.00\n", + " self.gae_param = 0.95\n", + " self.clip = 0.2 # epsilon from eq 7, default 0.2\n", + " self.ent_coeff = 0.\n", + " self.num_epoch = 10\n", + " self.num_steps = 2048\n", + " self.time_horizon = 2000000\n", + " self.max_episode_length = 10000\n", + " self.seed = 1\n", + "\n", + "params = Params()\n", + "\n", + "save_path= 'outputs/agent_portfolio-ddpo/{}_seperate_weights.pickle'.format('2017-07-21')\n", + "try:\n", + " os.makedirs(os.path.dirname(save_path))\n", + "except OSError:\n", + " pass\n", + "save_path" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Memory\n", + "refs\n", + "- https://github.com/openai/baselines/blob/master/baselines/deepq/replay_buffer.py\n", + "- https://github.com/jaara/AI-blog/blob/master/Seaquest-DDQN-PER.py" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:12.478946Z", + "start_time": "2017-08-05T14:42:12.411715Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "class ReplayMemory(object):\n", + " def __init__(self, capacity):\n", + " self.capacity = capacity\n", + " self.memory = []\n", + "\n", + " def push(self, events):\n", + " for event in zip(*events):\n", + " self.memory.append(event)\n", + " if len(self.memory)>self.capacity:\n", + " del self.memory[0]\n", + "\n", + " def clear(self):\n", + " self.memory = []\n", + "\n", + " def sample(self, batch_size):\n", + " samples = zip(*random.sample(self.memory, batch_size))\n", + " return map(lambda x: torch.cat(x, 0), samples)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:12.658795Z", + "start_time": "2017-08-05T14:42:12.483362Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "import torch.utils.data.sampler\n", + "\n", + "class PrioritisedReplayMemory(object):\n", + " def __init__(self, capacity):\n", + " self.capacity = capacity\n", + " self.memory = []\n", + "\n", + " def push(self, events):\n", + " # event is [states, actions, returns, advantages]\n", + " # [1x3x5x50,1x6,1x1,1x1]\n", + " for event in zip(*events):\n", + " self.memory.append(event)\n", + " if len(self.memory)>self.capacity:\n", + " del self.memory[0]\n", + "\n", + " def clear(self):\n", + " self.memory = []\n", + "\n", + " def sample(self, batch_size, beta=0.5):\n", + " \"\"\"\n", + " Take a weighted sample based on advantages.\n", + " \n", + " Half hearted implementation of algorithm 1 from https://arxiv.org/pdf/1511.05952.pdf\n", + " \n", + " Better one here https://github.com/openai/baselines/blob/master/baselines/deepq/replay_buffer.py#L157\n", + " \n", + " Parameters\n", + " ----------\n", + " batch_size: int\n", + " How many transitions to sample.\n", + " beta: float\n", + " To what degree to use importance weights\n", + " (0 - minimal corrections, 1 - full correction)\n", + " \n", + " Returns\n", + " ----------\n", + " states\n", + " actions\n", + " returns\n", + " advantages\n", + " \n", + " \"\"\"\n", + " # Sample transition\n", + " # we want to minimise loss so\n", + " # weights ~ -loss (bigger the weight when the loss is lower)\n", + " # and\n", + " # loss ~= -batch_advantages\n", + " # so \n", + " # weights ~ batch_advantages\n", + " ps1 = torch.FloatTensor([m[-1].squeeze().data[0] for m in self.memory]) \n", + " ps1 -= ps1.min()\n", + " ps1 /= ps1.sum()\n", + " \n", + " # batch_returns\n", + "# ps2 = torch.FloatTensor([m[-2].squeeze().data[0] for m in self.memory]) \n", + "# ps2 -= ps2.min()\n", + "# ps2 /= ps2.sum()\n", + " \n", + " ps = ps1#-ps2 # priority \n", + " ps -= ps.min() - 1.0\n", + " P = ps/ps.sum() # normalize\n", + " \n", + " # Compute importance-sampling weight\n", + " P = (len(P) * P) ** (-beta)\n", + " w = P / P.max()\n", + " \n", + " # to list and remove nans\n", + " w = w.numpy()\n", + "# w[np.isfinite(w)==False] = 0\n", + " w = w.tolist()\n", + " \n", + " idxs = torch.utils.data.sampler.WeightedRandomSampler(w, params.batch_size, replacement=False)\n", + " \n", + " # concatenate it into batches\n", + " samples = zip(*[self.memory[n] for n in idxs]) \n", + " return map(lambda x: torch.cat(x, 0), samples)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-04T12:46:15.785326Z", + "start_time": "2017-08-04T20:46:15.643789+08:00" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:12.685164Z", + "start_time": "2017-08-05T14:42:12.663278Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "# # Test by making sure the sample has a higher returns than the mean\n", + "# batch_states, batch_actions, batch_returns, batch_advantages = memory.sample(200)\n", + "\n", + "# ps1 = torch.FloatTensor([m[-1].squeeze().data[0] for m in memory.memory]) \n", + "# print(batch_advantages.mean(),ps1.mean())\n", + "# assert batch_advantages.mean().data[0]>ps1.mean()\n", + "\n", + "# # ps2 = torch.FloatTensor([m[-2].squeeze().data[0] for m in memory.memory]) \n", + "# # assert batch_returns.mean().data[0]>ps2.mean()\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-02T00:55:29.885772Z", + "start_time": "2017-08-02T08:55:29.883459+08:00" + } + }, + "source": [ + "# Enviroment" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:13.204219Z", + "start_time": "2017-08-05T14:42:12.686529Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "(3, 5, 50)" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from src.environments.portfolio import PortfolioEnv, sharpe, max_drawdown\n", + "\n", + "# we want to pemute the channels a little\n", + "\n", + "class PermutedPortfolioEnv(PortfolioEnv):\n", + " def reset(self, *args, **kwargs):\n", + " return np.transpose(super().reset(*args, **kwargs),(0,1,2))\n", + " def step(self, *args, **kwargs):\n", + " observation, reward, done, info = super().step(*args, **kwargs)\n", + " observation = np.transpose(observation,(2,0,1))\n", + " return observation, reward, done, info\n", + "\n", + "\n", + "df_train = pd.read_hdf('./data/poloniex_30m.hf',key='train')\n", + "env = PermutedPortfolioEnv(\n", + " df=df_train,\n", + " steps=128, \n", + " scale=True, \n", + " augment=0.00025, # let just overfit first,\n", + " trading_cost=0, #0.0025, # let just overfit first,\n", + " window_length = params.window_length, \n", + ")\n", + "env.seed(params.seed)\n", + "env.reset().shape\n", + "\n", + "df_test = pd.read_hdf('./data/poloniex_30m.hf',key='test')\n", + "env_test = PermutedPortfolioEnv(\n", + " df=df_test,\n", + " steps=128, \n", + " scale=True, \n", + " augment=0.00025, # let just overfit first,\n", + " trading_cost=0, #0.0025, # let just overfit first,\n", + " window_length = params.window_length, \n", + ")\n", + "env_test.seed(params.seed)\n", + "env_test.reset().shape" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-02T01:11:38.199434Z", + "start_time": "2017-08-02T09:11:38.155811+08:00" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:13.429447Z", + "start_time": "2017-08-05T14:42:13.335188Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "class GenericSharedModel(nn.Module):\n", + " def __init__(self, inputs, outputs):\n", + " super(GenericSharedModel, self).__init__()\n", + " num_inputs = int(np.prod(env.observation_space.shape))\n", + " num_outputs = int(np.prod(env.action_space.shape))\n", + " \n", + " # hidden layer sizes\n", + " h_size_1 = 128\n", + " h_size_2 = 128\n", + " \n", + " self.conv1 = nn.Conv2d(3, 2, (1, 3))\n", + " self.conv2 = nn.Conv2d(2, 20, (1, inputs[1] - 2))\n", + " \n", + " self.fc1 = nn.Linear(num_inputs, h_size_1)\n", + " self.fc2 = nn.Linear(h_size_1, h_size_2)\n", + " \n", + " self.mu = nn.Linear(h_size_2, num_outputs)\n", + " self.log_std = nn.Parameter(torch.zeros(num_outputs))\n", + " \n", + " \n", + " self.fc1b = nn.Linear(num_inputs, h_size_1)\n", + " self.fc2b = nn.Linear(h_size_1, h_size_2)\n", + " \n", + " self.v = nn.Linear(h_size_2,1)\n", + " \n", + " for name, p in self.named_parameters():\n", + " # init parameters\n", + " if 'bias' in name:\n", + " p.data.fill_(0)\n", + " '''\n", + " if 'mu.weight' in name:\n", + " p.data.normal_()\n", + " p.data /= torch.sum(p.data**2,0).expand_as(p.data)'''\n", + " \n", + " # mode\n", + " self.train()\n", + "\n", + " def forward(self, inputs):\n", + " # flatten\n", + " inputs = inputs.view((inputs.size()[0],-1))\n", + " x = F.relu(self.conv1(inputs))\n", + " x = F.relu(self.conv2(x))\n", + " h = x.view(x.size(0),-1) # Flatten\n", + " \n", + " # actor\n", + " x = F.tanh(self.fc1(h))\n", + " x = F.tanh(self.fc2(x))\n", + " \n", + " # the action\n", + " mu = F.softmax(self.mu(x))\n", + " \n", + " # exploration multiplier\n", + " log_std = F.sigmoid(torch.exp(self.log_std).unsqueeze(0).expand_as(mu))\n", + " \n", + " # critic\n", + " x = F.tanh(self.fc1(h))\n", + " x = F.tanh(self.fc2(x))\n", + " v = self.v(x)\n", + " return mu, log_std, v" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Train" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:13.455681Z", + "start_time": "2017-08-05T14:42:13.433918Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "def mkdir(base, name):\n", + " path = os.path.join(base, name)\n", + " if not os.path.exists(path):\n", + " os.makedirs(path)\n", + " return path" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:13.509000Z", + "start_time": "2017-08-05T14:42:13.460149Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "\n", + "# class Shared_grad_buffers():\n", + "# def __init__(self, model):\n", + "# self.grads = {}\n", + "# for name, p in model.named_parameters():\n", + "# self.grads[name+'_grad'] = torch.ones(p.size()).share_memory_()\n", + "\n", + "# def add_gradient(self, model):\n", + "# for name, p in model.named_parameters():\n", + "# self.grads[name+'_grad'] += p.grad.data\n", + "\n", + "# def reset(self):\n", + "# for name,grad in self.grads.items():\n", + "# self.grads[name].fill_(0)\n", + "\n", + "class Shared_obs_stats():\n", + " \"\"\"Like batchnorm for input data\"\"\"\n", + " def __init__(self, num_inputs):\n", + " self.n = torch.zeros(num_inputs).share_memory_()\n", + " self.mean = torch.zeros(num_inputs).share_memory_()\n", + " self.mean_diff = torch.zeros(num_inputs).share_memory_()\n", + " self.var = torch.zeros(num_inputs).share_memory_()\n", + "\n", + " def observes(self, obs):\n", + " # observation mean var updates\n", + " x = obs.data.squeeze()\n", + " self.n += 1.\n", + " last_mean = self.mean.clone()\n", + " self.mean += (x-self.mean)/self.n\n", + " self.mean_diff += (x-last_mean)*(x-self.mean)\n", + " self.var = torch.clamp(self.mean_diff/self.n, min=1e-2)\n", + "\n", + " def normalize(self, inputs):\n", + " obs_mean = Variable(self.mean.unsqueeze(0).expand_as(inputs))\n", + " obs_std = Variable(torch.sqrt(self.var).unsqueeze(0).expand_as(inputs))\n", + " return torch.clamp((inputs-obs_mean)/obs_std, -5., 5.)" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:13.550899Z", + "start_time": "2017-08-05T14:42:13.513430Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [ + "def normal(x, mu, sigma_sq):\n", + " a = (-1*(x-mu).pow(2)/(2*sigma_sq)).exp()\n", + " b = 1/(2*sigma_sq*np.pi).sqrt()\n", + " return a*b" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T14:42:13.633591Z", + "start_time": "2017-08-05T14:42:13.555326Z" + } + }, + "outputs": [ + { + "data": { + "text/plain": [ + "EIIE_CNN (\n", + " (conv1): Conv2d(3, 2, kernel_size=(1, 3), stride=(1, 1))\n", + " (conv2): Conv2d(2, 20, kernel_size=(1, 48), stride=(1, 1))\n", + " (conv3): Conv2d(20, 1, kernel_size=(1, 1), stride=(1, 1))\n", + " (head): Linear (5 -> 6)\n", + " (conv2b): Conv2d(2, 20, kernel_size=(1, 48), stride=(1, 1))\n", + " (conv3b): Conv2d(20, 1, kernel_size=(1, 1), stride=(1, 1))\n", + " (headb): Linear (5 -> 6)\n", + " (v): Linear (6 -> 1)\n", + ")" + ] + }, + "execution_count": 14, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "\n", + "cuda = False\n", + "torch.manual_seed(params.seed)\n", + "work_dir = mkdir('exp', 'ppo')\n", + "monitor_dir = mkdir(work_dir, 'monitor')\n", + "\n", + "# env = gym.make(params.env_name)\n", + "#env = wrappers.Monitor(env, monitor_dir, force=True)\n", + "\n", + "num_inputs = env.observation_space.shape[0]\n", + "num_outputs = env.action_space.shape[0]\n", + "\n", + "\n", + "#initialize network and optimizer\n", + "Model = GenericSharedModel\n", + "Model = EIIE_CNN\n", + "model = Model(env.observation_space.shape, env.action_space.shape)\n", + "if cuda: model.cuda()\n", + "\n", + "# shared_obs_stats = Shared_obs_stats(num_inputs)\n", + "optimizer = optim.Adam(model.parameters(), lr=params.lr)\n", + "model" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-04T02:50:27.128726Z", + "start_time": "2017-08-04T10:50:01.031402+08:00" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T20:16:27.987057Z", + "start_time": "2017-08-05T14:42:13.635113Z" + } + }, + "outputs": [ + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "376d3e3b25df4e2d9000dd4d8343b594" + } + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.049, cash_bias=0.3195, market_value=1.025, reward=1.138e-05, episode=15, loss=0.000261\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/wassname/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/torch/serialization.py:147: UserWarning: Couldn't retrieve source code for container of type EIIE_CNN. It won't be checked for correctness upon loading.\n", + " \"type \" + obj.__name__ + \". It won't be checked \"\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.073, cash_bias=0.157, market_value=1.048, reward=4.855e-06, episode=31, loss=0.0002609\n", + "portfolio_value=1.061, cash_bias=0.2, market_value=1.029, reward=-2.669e-05, episode=47, loss=0.000261\n", + "portfolio_value=1.042, cash_bias=0.295, market_value=1.03, reward=1.347e-05, episode=63, loss=0.0002603\n", + "portfolio_value=1.08, cash_bias=0.2887, market_value=1.066, reward=-1.589e-06, episode=79, loss=0.0002598\n", + "portfolio_value=1.005, cash_bias=0.2187, market_value=1.039, reward=-2.089e-05, episode=95, loss=0.0002598\n", + "portfolio_value=1.041, cash_bias=0.2338, market_value=1.057, reward=3.306e-05, episode=111, loss=0.0002597\n", + "portfolio_value=1.079, cash_bias=0.2115, market_value=1.035, reward=2.707e-05, episode=127, loss=0.0002596\n", + "portfolio_value=1.052, cash_bias=0.2053, market_value=1.03, reward=4.84e-06, episode=143, loss=0.0002596\n", + "portfolio_value=1.054, cash_bias=0.2824, market_value=1.069, reward=3.712e-06, episode=159, loss=0.0002596\n", + "portfolio_value=1.031, cash_bias=0.2244, market_value=1.045, reward=-1.331e-05, episode=175, loss=0.0002597\n", + "portfolio_value=1.075, cash_bias=0.2239, market_value=1.052, reward=7.596e-06, episode=191, loss=0.0002596\n", + "portfolio_value=1.043, cash_bias=0.2541, market_value=1.043, reward=1.193e-05, episode=207, loss=0.0002597\n", + "portfolio_value=1.034, cash_bias=0.1638, market_value=1.028, reward=1.954e-05, episode=223, loss=0.0002597\n", + "portfolio_value=1.042, cash_bias=0.209, market_value=1.037, reward=-8.082e-06, episode=239, loss=0.0002597\n", + "portfolio_value=1.071, cash_bias=0.1578, market_value=1.052, reward=-5.499e-06, episode=255, loss=0.0002596\n", + "portfolio_value=1.022, cash_bias=0.2591, market_value=1.012, reward=-3.455e-06, episode=271, loss=0.0002597\n", + "portfolio_value=1.01, cash_bias=0.3475, market_value=1.009, reward=4.03e-06, episode=287, loss=0.0002597\n", + "portfolio_value=1.012, cash_bias=0.3971, market_value=1.014, reward=-3.128e-06, episode=303, loss=0.0002597\n", + "portfolio_value=1.084, cash_bias=0.211, market_value=1.041, reward=1.116e-05, episode=319, loss=0.0002596\n", + "portfolio_value=1.022, cash_bias=0.2573, market_value=1.031, reward=-3.899e-06, episode=335, loss=0.0002597\n", + "portfolio_value=1.019, cash_bias=0.3925, market_value=1.022, reward=3.561e-05, episode=351, loss=0.0002597\n", + "portfolio_value=1.078, cash_bias=0.3282, market_value=1.039, reward=-2.892e-05, episode=367, loss=0.0002597\n", + "portfolio_value=1.022, cash_bias=0.3064, market_value=0.9923, reward=9.717e-06, episode=383, loss=0.0002597\n", + "portfolio_value=1.061, cash_bias=0.2684, market_value=1.044, reward=-3.57e-05, episode=399, loss=0.0002597\n", + "portfolio_value=1.042, cash_bias=0.2703, market_value=1.031, reward=2.17e-06, episode=415, loss=0.0002597\n", + "portfolio_value=1.044, cash_bias=0.2228, market_value=1.046, reward=2.258e-05, episode=431, loss=0.0002597\n", + "portfolio_value=1.044, cash_bias=0.3, market_value=1.05, reward=-4.255e-06, episode=447, loss=0.0002596\n", + "portfolio_value=1.047, cash_bias=0.1665, market_value=1.031, reward=-8.877e-07, episode=463, loss=0.0002597\n", + "portfolio_value=1.07, cash_bias=0.2616, market_value=1.055, reward=1.188e-05, episode=479, loss=0.0002596\n", + "portfolio_value=1.046, cash_bias=0.2893, market_value=1.045, reward=-5.866e-06, episode=495, loss=0.0002596\n", + "portfolio_value=1.003, cash_bias=0.2758, market_value=1.011, reward=3.015e-05, episode=511, loss=0.0002597\n", + "portfolio_value=1.063, cash_bias=0.06662, market_value=1.052, reward=9.413e-06, episode=527, loss=0.0002596\n", + "portfolio_value=1.01, cash_bias=0.2169, market_value=1.006, reward=-1.382e-05, episode=543, loss=0.0002597\n", + "portfolio_value=1.035, cash_bias=0.2092, market_value=1.037, reward=-2.035e-05, episode=559, loss=0.0002597\n", + "portfolio_value=1.022, cash_bias=0.0965, market_value=1.045, reward=-2.225e-05, episode=575, loss=0.0002597\n", + "portfolio_value=1.048, cash_bias=0.2815, market_value=1.045, reward=5.092e-06, episode=591, loss=0.0002597\n", + "portfolio_value=0.9595, cash_bias=0.2037, market_value=0.9956, reward=1.473e-07, episode=607, loss=0.0002598\n", + "portfolio_value=1.056, cash_bias=0.2607, market_value=1.026, reward=-1.019e-05, episode=623, loss=0.0002597\n", + "portfolio_value=1.016, cash_bias=0.2279, market_value=1.043, reward=-2.514e-05, episode=639, loss=0.0002597\n", + "portfolio_value=1.046, cash_bias=0.1889, market_value=1.025, reward=-4.179e-06, episode=655, loss=0.0002597\n", + "portfolio_value=1.019, cash_bias=0.2498, market_value=1.009, reward=-1.666e-05, episode=671, loss=0.0002597\n", + "portfolio_value=1.003, cash_bias=0.2711, market_value=1.023, reward=-5.755e-06, episode=687, loss=0.0002597\n", + "portfolio_value=0.979, cash_bias=0.2435, market_value=1.006, reward=7.097e-06, episode=703, loss=0.0002597\n", + "portfolio_value=1.066, cash_bias=0.3112, market_value=1.065, reward=-7.831e-06, episode=719, loss=0.0002596\n", + "portfolio_value=1.011, cash_bias=0.2126, market_value=1.032, reward=1.611e-06, episode=735, loss=0.0002597\n", + "portfolio_value=1.073, cash_bias=0.2961, market_value=1.046, reward=4.379e-06, episode=751, loss=0.0002596\n", + "portfolio_value=1.005, cash_bias=0.2581, market_value=1.004, reward=-8.174e-06, episode=767, loss=0.0002597\n", + "portfolio_value=0.9791, cash_bias=0.2135, market_value=1.016, reward=1.729e-05, episode=783, loss=0.0002597\n", + "portfolio_value=1.063, cash_bias=0.2582, market_value=1.052, reward=9.938e-06, episode=799, loss=0.0002596\n", + "portfolio_value=1.085, cash_bias=0.2437, market_value=1.087, reward=1.799e-05, episode=815, loss=0.0002596\n", + "portfolio_value=1.038, cash_bias=0.3149, market_value=1.039, reward=-8.042e-06, episode=831, loss=0.0002597\n", + "portfolio_value=1.025, cash_bias=0.2132, market_value=1.041, reward=1.322e-05, episode=847, loss=0.0002597\n", + "portfolio_value=1.054, cash_bias=0.2469, market_value=1.017, reward=-2.245e-05, episode=863, loss=0.0002597\n", + "portfolio_value=1.019, cash_bias=0.2093, market_value=1.013, reward=-4.591e-06, episode=879, loss=0.0002597\n", + "portfolio_value=1.048, cash_bias=0.2336, market_value=1.02, reward=-7.3e-06, episode=895, loss=0.0002597\n", + "portfolio_value= 1, cash_bias=0.1598, market_value=1.02, reward=-1.838e-05, episode=911, loss=0.0002597\n", + "portfolio_value=1.087, cash_bias=0.1657, market_value=1.039, reward=-4.382e-05, episode=927, loss=0.0002596\n", + "portfolio_value=0.9778, cash_bias=0.2886, market_value=1.012, reward=3.349e-05, episode=943, loss=0.0002598\n", + "portfolio_value=1.036, cash_bias=0.2306, market_value=1.03, reward=1.465e-05, episode=959, loss=0.0002597\n", + "portfolio_value=1.025, cash_bias=0.2092, market_value=1.034, reward=5.647e-06, episode=975, loss=0.0002597\n", + "portfolio_value=1.06, cash_bias=0.1236, market_value=1.046, reward=-3.17e-05, episode=991, loss=0.0002597\n", + "portfolio_value=1.054, cash_bias=0.3729, market_value=1.034, reward=-7.035e-06, episode=1007, loss=0.0002597\n", + "portfolio_value=1.029, cash_bias=0.2518, market_value=1.042, reward=7.582e-06, episode=1023, loss=0.0002596\n", + "portfolio_value=1.039, cash_bias=0.2861, market_value=1.015, reward=-3.131e-05, episode=1039, loss=0.0002597\n", + "portfolio_value=1.063, cash_bias=0.234, market_value=1.046, reward=6.244e-06, episode=1055, loss=0.0002596\n", + "portfolio_value=1.034, cash_bias=0.3041, market_value=1.027, reward=2.599e-05, episode=1071, loss=0.0002597\n", + "portfolio_value=1.016, cash_bias=0.3085, market_value=1.032, reward=1.614e-07, episode=1087, loss=0.0002597\n", + "portfolio_value=1.03, cash_bias=0.188, market_value=1.025, reward=-2.622e-05, episode=1103, loss=0.0002597\n", + "portfolio_value=1.125, cash_bias=0.2317, market_value=1.07, reward=3.869e-05, episode=1119, loss=0.0002595\n", + "portfolio_value=1.015, cash_bias=0.3645, market_value=1.029, reward=-1.668e-05, episode=1135, loss=0.0002597\n", + "portfolio_value=1.046, cash_bias=0.222, market_value=1.029, reward=5.882e-06, episode=1151, loss=0.0002597\n", + "portfolio_value=1.003, cash_bias=0.3438, market_value=1.01, reward=7.846e-08, episode=1167, loss=0.0002597\n", + "portfolio_value=1.023, cash_bias=0.1371, market_value=1.02, reward=6.101e-06, episode=1183, loss=0.0002597\n", + "portfolio_value=1.078, cash_bias=0.1762, market_value=1.049, reward=8.5e-07, episode=1199, loss=0.0002596\n", + "portfolio_value=1.045, cash_bias=0.1546, market_value=1.029, reward=-3.091e-05, episode=1215, loss=0.0002597\n", + "portfolio_value=1.017, cash_bias=0.2206, market_value=1.008, reward=1.839e-06, episode=1231, loss=0.0002597\n", + "portfolio_value=1.037, cash_bias=0.2597, market_value=1.036, reward=7.963e-06, episode=1247, loss=0.0002597\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.007, cash_bias=0.2388, market_value=1.022, reward=-2.415e-05, episode=1263, loss=0.0002598\n", + "portfolio_value=1.04, cash_bias=0.3485, market_value=1.05, reward=7.861e-06, episode=1279, loss=0.0002596\n", + "portfolio_value=1.016, cash_bias=0.2901, market_value=1.054, reward=-6.921e-06, episode=1295, loss=0.0002597\n", + "portfolio_value=1.064, cash_bias=0.2432, market_value=1.036, reward=1.058e-05, episode=1311, loss=0.0002596\n", + "portfolio_value=1.026, cash_bias=0.342, market_value=0.9962, reward=-2.477e-05, episode=1327, loss=0.0002597\n", + "portfolio_value=0.999, cash_bias=0.3548, market_value=1.017, reward=1.143e-05, episode=1343, loss=0.0002597\n", + "portfolio_value=1.083, cash_bias=0.3556, market_value=1.067, reward=4.655e-05, episode=1359, loss=0.0002596\n", + "portfolio_value=0.9973, cash_bias=0.212, market_value=1.014, reward=-1.284e-05, episode=1375, loss=0.0002597\n", + "portfolio_value=1.02, cash_bias=0.4374, market_value=1.038, reward=-8.869e-06, episode=1391, loss=0.0002597\n", + "portfolio_value=1.05, cash_bias=0.2263, market_value=1.039, reward=-9.761e-06, episode=1407, loss=0.0002597\n", + "portfolio_value=1.023, cash_bias=0.3083, market_value=1.04, reward=7.977e-06, episode=1423, loss=0.0002597\n", + "portfolio_value=1.062, cash_bias=0.2123, market_value=1.04, reward=3.636e-05, episode=1439, loss=0.0002596\n", + "portfolio_value=1.063, cash_bias=0.2562, market_value=1.046, reward=1.116e-05, episode=1455, loss=0.0002597\n", + "portfolio_value=1.089, cash_bias=0.1512, market_value=1.034, reward=2.528e-06, episode=1471, loss=0.0002596\n", + "portfolio_value=1.082, cash_bias=0.1209, market_value=1.044, reward=-1.58e-06, episode=1487, loss=0.0002596\n", + "portfolio_value=1.041, cash_bias=0.1669, market_value=1.072, reward=1.56e-05, episode=1503, loss=0.0002597\n", + "portfolio_value=1.054, cash_bias=0.1775, market_value=1.039, reward=1.24e-05, episode=1519, loss=0.0002597\n", + "portfolio_value=1.053, cash_bias=0.1769, market_value=1.072, reward=-1.407e-05, episode=1535, loss=0.0002597\n", + "portfolio_value=1.058, cash_bias=0.192, market_value=1.052, reward=1.261e-06, episode=1551, loss=0.0002597\n", + "portfolio_value=1.078, cash_bias=0.2238, market_value=1.057, reward=2.148e-05, episode=1567, loss=0.0002596\n", + "portfolio_value=0.9741, cash_bias=0.2897, market_value=0.9993, reward=-3.113e-06, episode=1583, loss=0.0002597\n", + "portfolio_value=1.014, cash_bias=0.1276, market_value=1.013, reward=8.929e-06, episode=1599, loss=0.0002597\n", + "portfolio_value=1.026, cash_bias=0.3655, market_value=1.005, reward=2.139e-05, episode=1615, loss=0.0002597\n", + "portfolio_value=1.055, cash_bias=0.2249, market_value=1.091, reward=1.691e-05, episode=1631, loss=0.0002597\n", + "portfolio_value=1.066, cash_bias=0.2215, market_value=1.059, reward=3.05e-06, episode=1647, loss=0.0002597\n", + "portfolio_value=1.027, cash_bias=0.1846, market_value=1.021, reward=5.417e-06, episode=1663, loss=0.0002597\n", + "portfolio_value=1.022, cash_bias=0.381, market_value=1.032, reward=-8.426e-06, episode=1679, loss=0.0002597\n", + "portfolio_value=0.9574, cash_bias=0.2881, market_value=1.017, reward=-2.511e-05, episode=1695, loss=0.0002597\n", + "portfolio_value=0.9886, cash_bias=0.2447, market_value=1.01, reward=1.313e-05, episode=1711, loss=0.0002597\n", + "portfolio_value=0.9792, cash_bias=0.2562, market_value=1.009, reward=-2.441e-06, episode=1727, loss=0.0002598\n", + "portfolio_value=1.038, cash_bias=0.09988, market_value=1.042, reward=-8.703e-06, episode=1743, loss=0.0002597\n", + "portfolio_value=1.074, cash_bias=0.2846, market_value=1.059, reward=2.865e-05, episode=1759, loss=0.0002596\n", + "portfolio_value=1.023, cash_bias=0.274, market_value=1.05, reward=-2.552e-05, episode=1775, loss=0.0002597\n", + "portfolio_value=1.069, cash_bias=0.2899, market_value=1.052, reward=1.134e-05, episode=1791, loss=0.0002596\n", + "portfolio_value=0.9911, cash_bias=0.146, market_value=0.9987, reward=6.595e-06, episode=1807, loss=0.0002597\n", + "portfolio_value=1.067, cash_bias=0.1947, market_value=1.047, reward=-6.761e-07, episode=1823, loss=0.0002597\n", + "portfolio_value=1.034, cash_bias=0.2846, market_value=1.039, reward=2.492e-05, episode=1839, loss=0.0002597\n", + "portfolio_value=1.075, cash_bias=0.199, market_value=1.071, reward=2.324e-05, episode=1855, loss=0.0002596\n", + "portfolio_value=1.075, cash_bias=0.3307, market_value=1.058, reward=8.866e-06, episode=1871, loss=0.0002597\n", + "portfolio_value=0.9926, cash_bias=0.4017, market_value=1.008, reward=-3.466e-06, episode=1887, loss=0.0002597\n", + "portfolio_value=1.026, cash_bias=0.3285, market_value=1.021, reward=7.626e-06, episode=1903, loss=0.0002597\n", + "portfolio_value=1.028, cash_bias=0.1264, market_value=1.012, reward=-7.278e-06, episode=1919, loss=0.0002597\n", + "portfolio_value=1.107, cash_bias=0.2651, market_value=1.055, reward=-6.759e-06, episode=1935, loss=0.0002596\n", + "portfolio_value=1.02, cash_bias=0.1309, market_value=1.016, reward=-9.021e-07, episode=1951, loss=0.0002597\n", + "portfolio_value=0.9974, cash_bias=0.111, market_value=1.011, reward=6.877e-06, episode=1967, loss=0.0002597\n", + "portfolio_value=1.07, cash_bias=0.2467, market_value=1.051, reward=-1.74e-05, episode=1983, loss=0.0002596\n", + "portfolio_value=1.053, cash_bias=0.3121, market_value=1.031, reward=-1.352e-05, episode=1999, loss=0.0002597\n", + "portfolio_value=1.053, cash_bias=0.0935, market_value=1.042, reward=3.954e-06, episode=2015, loss=0.0002597\n", + "portfolio_value=1.047, cash_bias=0.2502, market_value=1.047, reward=-3.634e-06, episode=2031, loss=0.0002596\n", + "portfolio_value=1.02, cash_bias=0.2918, market_value=1.015, reward=-1.002e-05, episode=2047, loss=0.0002597\n", + "portfolio_value=1.051, cash_bias=0.1352, market_value=1.036, reward=-4.293e-06, episode=2063, loss=0.0002597\n", + "portfolio_value=1.054, cash_bias=0.09199, market_value=1.023, reward=2.045e-08, episode=2079, loss=0.0002596\n", + "portfolio_value=1.064, cash_bias=0.1685, market_value=1.065, reward=2.063e-05, episode=2095, loss=0.0002597\n", + "portfolio_value=1.045, cash_bias=0.162, market_value=1.042, reward=-1.358e-05, episode=2111, loss=0.0002596\n", + "portfolio_value=1.033, cash_bias=0.3105, market_value=1.003, reward=1.137e-05, episode=2127, loss=0.0002597\n", + "portfolio_value=1.034, cash_bias=0.1983, market_value=1.017, reward=1.617e-05, episode=2143, loss=0.0002597\n", + "portfolio_value=1.014, cash_bias=0.1801, market_value=1.035, reward=-3.807e-06, episode=2159, loss=0.0002597\n", + "portfolio_value=1.078, cash_bias=0.2546, market_value=1.035, reward=-1.547e-05, episode=2175, loss=0.0002596\n", + "portfolio_value=1.056, cash_bias=0.2306, market_value=1.039, reward=-1.225e-05, episode=2191, loss=0.0002597\n", + "portfolio_value=1.028, cash_bias=0.2872, market_value=1.035, reward=1.61e-05, episode=2207, loss=0.0002596\n", + "portfolio_value=1.074, cash_bias=0.2505, market_value=1.05, reward=4.188e-06, episode=2223, loss=0.0002596\n", + "portfolio_value=1.02, cash_bias=0.2327, market_value=1.033, reward=5.353e-07, episode=2239, loss=0.0002596\n", + "portfolio_value=1.065, cash_bias=0.2019, market_value=1.043, reward=-7.215e-06, episode=2255, loss=0.0002597\n", + "portfolio_value=1.047, cash_bias=0.1574, market_value=1.056, reward=2.824e-08, episode=2271, loss=0.0002597\n", + "portfolio_value=1.051, cash_bias=0.2043, market_value=1.023, reward=-9.474e-06, episode=2287, loss=0.0002597\n", + "portfolio_value=1.02, cash_bias=0.2678, market_value=1.045, reward=2.015e-05, episode=2303, loss=0.0002597\n", + "portfolio_value=1.023, cash_bias=0.2845, market_value=1.038, reward=2.081e-05, episode=2319, loss=0.0002597\n", + "portfolio_value=1.076, cash_bias=0.1268, market_value=1.056, reward=7.156e-06, episode=2335, loss=0.0002597\n", + "portfolio_value=0.9862, cash_bias=0.3342, market_value=1.009, reward=-6.088e-06, episode=2351, loss=0.0002597\n", + "portfolio_value=1.045, cash_bias=0.1857, market_value=1.037, reward=1.089e-05, episode=2367, loss=0.0002596\n", + "portfolio_value=1.058, cash_bias=0.2483, market_value=1.054, reward=3.464e-06, episode=2383, loss=0.0002596\n", + "portfolio_value=1.075, cash_bias=0.2803, market_value=1.051, reward=1.214e-05, episode=2399, loss=0.0002596\n", + "portfolio_value=1.029, cash_bias=0.3558, market_value=1.039, reward=9.841e-06, episode=2415, loss=0.0002596\n", + "portfolio_value=1.038, cash_bias=0.2801, market_value=1.022, reward=-4.39e-06, episode=2431, loss=0.0002597\n", + "portfolio_value=1.021, cash_bias=0.1814, market_value=1.035, reward=-8.762e-06, episode=2447, loss=0.0002596\n", + "portfolio_value=1.115, cash_bias=0.2365, market_value=1.061, reward=-3.506e-06, episode=2463, loss=0.0002596\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.082, cash_bias=0.2483, market_value=1.049, reward=-4.433e-06, episode=2479, loss=0.0002596\n", + "portfolio_value=1.077, cash_bias=0.1755, market_value=1.045, reward=2.396e-05, episode=2495, loss=0.0002597\n", + "portfolio_value=1.054, cash_bias=0.1596, market_value=1.034, reward=6.848e-06, episode=2511, loss=0.0002597\n", + "portfolio_value=0.9937, cash_bias=0.2124, market_value=1.021, reward=1.722e-06, episode=2527, loss=0.0002598\n", + "portfolio_value=1.022, cash_bias=0.2606, market_value=1.013, reward=1.049e-05, episode=2543, loss=0.0002597\n", + "portfolio_value=1.037, cash_bias=0.3168, market_value=1.039, reward=-1.62e-05, episode=2559, loss=0.0002597\n", + "portfolio_value=1.028, cash_bias=0.1606, market_value=1.03, reward=1.206e-05, episode=2575, loss=0.0002596\n", + "portfolio_value=1.035, cash_bias=0.1901, market_value=1.029, reward=-6.543e-06, episode=2591, loss=0.0002597\n", + "portfolio_value=1.044, cash_bias=0.1994, market_value=1.027, reward=-5.397e-06, episode=2607, loss=0.0002597\n", + "portfolio_value=1.047, cash_bias=0.2381, market_value=1.04, reward=3.636e-06, episode=2623, loss=0.0002596\n", + "portfolio_value=1.039, cash_bias=0.1273, market_value=1.029, reward=1.416e-06, episode=2639, loss=0.0002597\n", + "portfolio_value=1.065, cash_bias=0.3092, market_value=1.057, reward=-3.576e-05, episode=2655, loss=0.0002597\n", + "portfolio_value=0.9902, cash_bias=0.3265, market_value=1.031, reward=-7.08e-06, episode=2671, loss=0.0002597\n", + "portfolio_value=1.017, cash_bias=0.1652, market_value=1.032, reward=-1.118e-05, episode=2687, loss=0.0002597\n", + "portfolio_value=1.037, cash_bias=0.1734, market_value=1.01, reward=-6.62e-06, episode=2703, loss=0.0002597\n", + "portfolio_value=0.9936, cash_bias=0.1831, market_value=1.007, reward=3.092e-05, episode=2719, loss=0.0002598\n", + "portfolio_value=1.022, cash_bias=0.1945, market_value=1.022, reward=-1.785e-05, episode=2735, loss=0.0002597\n", + "portfolio_value=1.033, cash_bias=0.282, market_value=1.028, reward=-1.638e-05, episode=2751, loss=0.0002597\n", + "portfolio_value=1.048, cash_bias=0.2153, market_value=1.047, reward=6.979e-07, episode=2767, loss=0.0002596\n", + "portfolio_value=1.073, cash_bias=0.236, market_value=1.046, reward=-1.331e-05, episode=2783, loss=0.0002596\n", + "portfolio_value=1.047, cash_bias=0.2982, market_value=1.034, reward=1.907e-05, episode=2799, loss=0.0002596\n", + "portfolio_value=1.013, cash_bias=0.1665, market_value=1.007, reward=-1.019e-05, episode=2815, loss=0.0002655\n", + "portfolio_value=1.052, cash_bias=0.3373, market_value=1.044, reward=1.295e-06, episode=2831, loss=0.0002925\n", + "portfolio_value=1.055, cash_bias=0.1715, market_value=1.051, reward=1.596e-06, episode=2847, loss=0.0003064\n", + "portfolio_value=1.073, cash_bias=0.2794, market_value=1.047, reward=3.216e-06, episode=2863, loss=0.0003113\n", + "portfolio_value=1.038, cash_bias=0.2094, market_value=1.051, reward=1.746e-06, episode=2879, loss=0.000313\n", + "portfolio_value=1.058, cash_bias=0.1451, market_value=1.036, reward=-2.404e-05, episode=2895, loss=0.0003136\n", + "portfolio_value=1.034, cash_bias=0.2145, market_value=1.029, reward=2.631e-05, episode=2911, loss=0.0003139\n", + "portfolio_value=1.056, cash_bias=0.109, market_value=1.039, reward=1.238e-05, episode=2927, loss=0.000314\n", + "portfolio_value=0.9901, cash_bias=0.1403, market_value=1.034, reward=-5.914e-06, episode=2943, loss=0.0003141\n", + "portfolio_value=1.045, cash_bias=0.2527, market_value=1.045, reward=-2.439e-05, episode=2959, loss=0.000314\n", + "portfolio_value=1.01, cash_bias=0.1973, market_value=1.022, reward=-1.527e-05, episode=2975, loss=0.000314\n", + "portfolio_value=1.015, cash_bias=0.1986, market_value=1.015, reward=-2.428e-05, episode=2991, loss=0.000314\n", + "portfolio_value=1.056, cash_bias=0.2596, market_value=1.038, reward=1.267e-05, episode=3007, loss=0.0003141\n", + "portfolio_value=1.069, cash_bias=0.2672, market_value=1.028, reward=5.371e-06, episode=3023, loss=0.000314\n", + "portfolio_value=0.9843, cash_bias=0.1858, market_value=1.009, reward=3.106e-05, episode=3039, loss=0.0003145\n", + "portfolio_value=1.05, cash_bias=0.1699, market_value=1.041, reward=-1.939e-05, episode=3055, loss=0.000314\n", + "portfolio_value=1.008, cash_bias=0.2046, market_value=1.017, reward=-8.649e-06, episode=3071, loss=0.0003141\n", + "portfolio_value=1.031, cash_bias=0.218, market_value=1.008, reward=-2.404e-05, episode=3087, loss=0.0003141\n", + "portfolio_value=1.004, cash_bias=0.2606, market_value=1.035, reward=1.415e-05, episode=3103, loss=0.000314\n", + "portfolio_value=1.033, cash_bias=0.1629, market_value=1.02, reward=3.285e-05, episode=3119, loss=0.000314\n", + "portfolio_value=1.056, cash_bias=0.238, market_value=1.03, reward=1.461e-05, episode=3135, loss=0.000314\n", + "portfolio_value=1.042, cash_bias=0.248, market_value=1.047, reward=3.818e-05, episode=3151, loss=0.000314\n", + "portfolio_value=1.062, cash_bias=0.2257, market_value=1.037, reward=4.211e-06, episode=3167, loss=0.000314\n", + "portfolio_value=1.075, cash_bias=0.2484, market_value=1.045, reward=6.419e-07, episode=3183, loss=0.0003139\n", + "portfolio_value=1.032, cash_bias=0.1695, market_value=1.005, reward=9.606e-06, episode=3199, loss=0.0003141\n", + "portfolio_value=0.9987, cash_bias=0.05394, market_value=0.9945, reward=1.452e-05, episode=3215, loss=0.0003141\n", + "portfolio_value=1.061, cash_bias=0.1841, market_value=1.069, reward=2.555e-05, episode=3231, loss=0.000314\n", + "portfolio_value=1.035, cash_bias=0.2008, market_value=1.034, reward=-4.94e-06, episode=3247, loss=0.000314\n", + "portfolio_value=1.016, cash_bias=0.1332, market_value=1.046, reward=-3.813e-06, episode=3263, loss=0.0003141\n", + "portfolio_value=1.03, cash_bias=0.2701, market_value=1.029, reward=6.834e-06, episode=3279, loss=0.000314\n", + "portfolio_value=1.034, cash_bias=0.1737, market_value=1.013, reward=5.826e-05, episode=3295, loss=0.000314\n", + "portfolio_value=1.024, cash_bias=0.3366, market_value=1.019, reward=7.789e-06, episode=3311, loss=0.0003126\n", + "portfolio_value=1.057, cash_bias=0.2828, market_value=1.039, reward=3.513e-06, episode=3327, loss=0.0002783\n", + "portfolio_value=1.014, cash_bias=0.2119, market_value=1.029, reward=-7.623e-06, episode=3343, loss=0.000255\n", + "portfolio_value=1.037, cash_bias=0.2616, market_value=1.034, reward=1.51e-05, episode=3359, loss=0.000247\n", + "portfolio_value=1.046, cash_bias=0.2213, market_value=1.04, reward=1.721e-05, episode=3375, loss=0.0002442\n", + "portfolio_value=1.08, cash_bias=0.1857, market_value=1.03, reward=-1.397e-05, episode=3391, loss=0.0002433\n", + "portfolio_value=1.006, cash_bias=0.2562, market_value=1.024, reward=2.624e-05, episode=3407, loss=0.000243\n", + "portfolio_value=1.07, cash_bias=0.3687, market_value=1.023, reward=-1.201e-06, episode=3423, loss=0.0002428\n", + "portfolio_value=1.04, cash_bias=0.2653, market_value=1.06, reward=-3.818e-06, episode=3439, loss=0.0002428\n", + "portfolio_value=1.046, cash_bias=0.1966, market_value=1.011, reward=4.979e-06, episode=3455, loss=0.0002428\n", + "portfolio_value=1.066, cash_bias=0.2229, market_value=1.063, reward=-2.646e-05, episode=3471, loss=0.0002427\n", + "portfolio_value=1.047, cash_bias=0.2385, market_value=1.024, reward=8.423e-06, episode=3487, loss=0.0002427\n", + "portfolio_value=1.049, cash_bias=0.2528, market_value=1.028, reward=-1.822e-05, episode=3503, loss=0.0002428\n", + "portfolio_value=1.06, cash_bias=0.1749, market_value=1.039, reward=1.28e-05, episode=3519, loss=0.0002428\n", + "portfolio_value=1.031, cash_bias=0.2156, market_value=1.034, reward=1.339e-05, episode=3535, loss=0.0002428\n", + "portfolio_value=1.041, cash_bias=0.2181, market_value=1.022, reward=2.225e-05, episode=3551, loss=0.0002428\n", + "portfolio_value=1.053, cash_bias=0.3002, market_value=1.067, reward=1.347e-05, episode=3567, loss=0.0002428\n", + "portfolio_value=1.035, cash_bias=0.169, market_value=1.036, reward=8.124e-06, episode=3583, loss=0.0002428\n", + "portfolio_value=1.09, cash_bias=0.2564, market_value=1.057, reward=7.005e-06, episode=3599, loss=0.0002428\n", + "portfolio_value=1.123, cash_bias=0.2, market_value=1.038, reward=-7.596e-06, episode=3615, loss=0.0002427\n", + "portfolio_value=1.065, cash_bias=0.2957, market_value=1.045, reward=8.197e-07, episode=3631, loss=0.0002427\n", + "portfolio_value=1.046, cash_bias=0.2336, market_value=1.052, reward=1.142e-05, episode=3647, loss=0.0002428\n", + "portfolio_value=1.031, cash_bias=0.2077, market_value=1.046, reward=5.337e-06, episode=3663, loss=0.0002428\n", + "portfolio_value=1.017, cash_bias=0.2524, market_value=1.022, reward=-5.951e-06, episode=3679, loss=0.0002428\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.04, cash_bias=0.2597, market_value=1.037, reward=-1.03e-05, episode=3695, loss=0.0002427\n", + "portfolio_value=0.9749, cash_bias=0.2041, market_value=1.004, reward=-1.762e-05, episode=3711, loss=0.0002428\n", + "portfolio_value=1.055, cash_bias=0.1816, market_value=1.049, reward=-2.367e-05, episode=3727, loss=0.0002432\n", + "portfolio_value=1.015, cash_bias=0.2677, market_value=1.011, reward=9.779e-06, episode=3743, loss=0.0002444\n", + "portfolio_value=1.015, cash_bias=0.2185, market_value=1.029, reward=2.224e-05, episode=3759, loss=0.0002457\n", + "portfolio_value=0.9814, cash_bias=0.3043, market_value=1.02, reward=-6.094e-05, episode=3775, loss=0.0002461\n", + "portfolio_value=1.069, cash_bias=0.3345, market_value=1.039, reward=-3.511e-05, episode=3791, loss=0.0002462\n", + "portfolio_value=1.048, cash_bias=0.2425, market_value=1.027, reward=1.796e-05, episode=3807, loss=0.0002463\n", + "portfolio_value=1.023, cash_bias=0.2241, market_value=1.015, reward=-3.637e-06, episode=3823, loss=0.0002463\n", + "portfolio_value=1.039, cash_bias=0.2803, market_value=1.048, reward=-2.135e-05, episode=3839, loss=0.0002463\n", + "portfolio_value=1.006, cash_bias=0.2142, market_value=1.023, reward=-2.222e-05, episode=3855, loss=0.0002464\n", + "portfolio_value=1.012, cash_bias=0.2175, market_value=1.056, reward=-9.604e-06, episode=3871, loss=0.0002463\n", + "portfolio_value=0.987, cash_bias=0.2014, market_value=1.026, reward=8.164e-06, episode=3887, loss=0.0002463\n", + "portfolio_value=1.072, cash_bias=0.1485, market_value=1.065, reward=-6.405e-06, episode=3903, loss=0.0002462\n", + "portfolio_value=1.037, cash_bias=0.1457, market_value=1.026, reward=2.276e-05, episode=3919, loss=0.0002463\n", + "portfolio_value=1.056, cash_bias=0.1908, market_value=1.044, reward=-1.506e-05, episode=3935, loss=0.0002463\n", + "portfolio_value=1.033, cash_bias=0.1386, market_value=1.023, reward=4.033e-05, episode=3951, loss=0.0002463\n", + "portfolio_value=0.9791, cash_bias=0.3151, market_value=1.03, reward=6.689e-05, episode=3967, loss=0.0002464\n", + "portfolio_value=1.025, cash_bias=0.1621, market_value=1.023, reward=1.659e-05, episode=3983, loss=0.0002463\n", + "portfolio_value=1.04, cash_bias=0.336, market_value=1.029, reward=-1.483e-05, episode=3999, loss=0.0002463\n", + "portfolio_value=1.039, cash_bias=0.1015, market_value=1.038, reward=1.214e-06, episode=4015, loss=0.0002462\n", + "portfolio_value=1.061, cash_bias=0.3907, market_value=1.065, reward=-1.059e-05, episode=4031, loss=0.0002463\n", + "portfolio_value=0.9948, cash_bias=0.2541, market_value=1.036, reward=1.355e-05, episode=4047, loss=0.0002463\n", + "portfolio_value=1.023, cash_bias=0.197, market_value=1.026, reward=-8.122e-06, episode=4063, loss=0.0002463\n", + "portfolio_value=1.015, cash_bias=0.1215, market_value=1.027, reward=-5.83e-06, episode=4079, loss=0.0002463\n", + "portfolio_value=1.058, cash_bias=0.2376, market_value=1.034, reward=8.745e-06, episode=4095, loss=0.0002463\n", + "portfolio_value=1.018, cash_bias=0.1526, market_value=1.003, reward=-7.922e-06, episode=4111, loss=0.000247\n", + "portfolio_value=1.03, cash_bias=0.2973, market_value=1.023, reward=-2.244e-05, episode=4127, loss=0.0002463\n", + "portfolio_value=1.013, cash_bias=0.2523, market_value=1.011, reward=-3.343e-06, episode=4143, loss=0.0002463\n", + "portfolio_value=1.028, cash_bias=0.1842, market_value=1.012, reward=1.888e-05, episode=4159, loss=0.0002463\n", + "portfolio_value=1.042, cash_bias=0.2446, market_value=1.04, reward=4.164e-06, episode=4175, loss=0.0002463\n", + "portfolio_value=1.041, cash_bias=0.214, market_value=1.037, reward=3.648e-05, episode=4191, loss=0.0002463\n", + "portfolio_value=1.051, cash_bias=0.2181, market_value=1.064, reward=-1.044e-06, episode=4207, loss=0.0002463\n", + "portfolio_value=1.045, cash_bias=0.1533, market_value=1.056, reward=4.78e-06, episode=4223, loss=0.0002463\n", + "portfolio_value=1.063, cash_bias=0.2234, market_value=1.062, reward=-1.309e-05, episode=4239, loss=0.0002462\n", + "portfolio_value=1.027, cash_bias=0.153, market_value=1.015, reward=-4.643e-05, episode=4255, loss=0.0002464\n", + "portfolio_value=1.056, cash_bias=0.2295, market_value=1.046, reward=-2.098e-05, episode=4271, loss=0.0002463\n", + "portfolio_value=1.079, cash_bias=0.2795, market_value=1.049, reward=-1.544e-05, episode=4287, loss=0.0002463\n", + "portfolio_value=1.078, cash_bias=0.1367, market_value=1.06, reward=1.502e-06, episode=4303, loss=0.0002462\n", + "portfolio_value=1.026, cash_bias=0.1059, market_value=1.027, reward=-1.913e-05, episode=4319, loss=0.0002463\n", + "portfolio_value=1.031, cash_bias=0.1995, market_value=1.023, reward=4.464e-05, episode=4335, loss=0.0002463\n", + "portfolio_value=1.006, cash_bias=0.3032, market_value=1.019, reward=2.026e-07, episode=4351, loss=0.0002463\n", + "portfolio_value=1.054, cash_bias=0.1241, market_value=1.044, reward=1.543e-05, episode=4367, loss=0.0002463\n", + "portfolio_value=1.024, cash_bias=0.08742, market_value=1.013, reward=1.941e-05, episode=4383, loss=0.0002463\n", + "portfolio_value=1.085, cash_bias=0.3353, market_value=1.071, reward=5.327e-05, episode=4399, loss=0.0002462\n", + "portfolio_value=0.9986, cash_bias=0.288, market_value=1.014, reward=-3.813e-06, episode=4415, loss=0.0002463\n", + "portfolio_value=1.06, cash_bias=0.2074, market_value=1.054, reward=-1.682e-05, episode=4431, loss=0.0002463\n", + "portfolio_value=1.083, cash_bias=0.4409, market_value=1.071, reward=-1.761e-05, episode=4447, loss=0.0002463\n", + "portfolio_value=1.056, cash_bias=0.1488, market_value=1.03, reward=3.009e-06, episode=4463, loss=0.0002462\n", + "portfolio_value=1.042, cash_bias=0.2071, market_value=1.027, reward=8.215e-06, episode=4479, loss=0.0002463\n", + "portfolio_value=0.9636, cash_bias=0.2135, market_value=1.028, reward=-2.877e-06, episode=4495, loss=0.0002463\n", + "portfolio_value=1.061, cash_bias=0.2229, market_value=1.052, reward=3.912e-06, episode=4511, loss=0.0002463\n", + "portfolio_value=1.072, cash_bias=0.1527, market_value=1.051, reward=4.169e-07, episode=4527, loss=0.0002462\n", + "portfolio_value=0.9841, cash_bias=0.323, market_value=1.014, reward=-5.045e-06, episode=4543, loss=0.0002463\n", + "portfolio_value=1.03, cash_bias=0.2262, market_value=1.041, reward=-1.183e-05, episode=4559, loss=0.0002463\n", + "portfolio_value=1.061, cash_bias=0.314, market_value=1.04, reward=-2.142e-05, episode=4575, loss=0.0002462\n", + "portfolio_value=1.016, cash_bias=0.2146, market_value=1.014, reward=7.488e-06, episode=4591, loss=0.0002463\n", + "portfolio_value=1.025, cash_bias=0.236, market_value=1.039, reward=-7.708e-06, episode=4607, loss=0.0002463\n", + "portfolio_value=0.9798, cash_bias=0.3238, market_value=1.018, reward=-1.656e-05, episode=4623, loss=0.0002467\n", + "portfolio_value=1.089, cash_bias=0.2815, market_value=1.07, reward=-4.722e-06, episode=4639, loss=0.0002463\n", + "portfolio_value=1.019, cash_bias=0.1315, market_value=1.037, reward=-8.833e-06, episode=4655, loss=0.0002463\n", + "portfolio_value=1.067, cash_bias=0.1567, market_value=1.04, reward=-1.755e-05, episode=4671, loss=0.0002462\n", + "portfolio_value=1.029, cash_bias=0.1944, market_value=1.015, reward=-4.246e-06, episode=4687, loss=0.0002463\n", + "portfolio_value=1.048, cash_bias=0.1103, market_value=1.024, reward=-1.025e-05, episode=4703, loss=0.0002463\n", + "portfolio_value=1.017, cash_bias=0.1066, market_value=1.001, reward=-5.37e-06, episode=4719, loss=0.0002463\n", + "portfolio_value=1.051, cash_bias=0.2939, market_value=1.025, reward=2.173e-05, episode=4735, loss=0.0002463\n", + "portfolio_value=1.005, cash_bias=0.2694, market_value=1.012, reward=6.258e-06, episode=4751, loss=0.0002463\n", + "portfolio_value=1.057, cash_bias=0.1043, market_value=1.049, reward=-3.095e-05, episode=4767, loss=0.0002463\n", + "portfolio_value=1.031, cash_bias=0.1843, market_value=1.024, reward=-2.252e-06, episode=4783, loss=0.0002463\n", + "portfolio_value=1.042, cash_bias=0.1863, market_value=1.033, reward=-4.956e-06, episode=4799, loss=0.0002463\n", + "portfolio_value=1.02, cash_bias=0.1539, market_value=1.056, reward=3.285e-05, episode=4815, loss=0.0002463\n", + "portfolio_value=1.004, cash_bias=0.2156, market_value=1.041, reward=9.886e-06, episode=4831, loss=0.0002463\n", + "portfolio_value=1.022, cash_bias=0.1431, market_value=1.006, reward=-1.188e-05, episode=4847, loss=0.0002463\n", + "portfolio_value=1.04, cash_bias=0.1393, market_value=1.033, reward=-5.914e-06, episode=4863, loss=0.0002463\n", + "portfolio_value=1.047, cash_bias=0.1701, market_value=1.056, reward=4.019e-05, episode=4879, loss=0.0002463\n", + "portfolio_value=1.008, cash_bias=0.1978, market_value=1.032, reward=1.884e-05, episode=4895, loss=0.0002463\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.024, cash_bias=0.1079, market_value=1.051, reward=9.701e-06, episode=4911, loss=0.0002462\n", + "portfolio_value=1.033, cash_bias=0.2285, market_value=1.033, reward=1.263e-05, episode=4927, loss=0.0002463\n", + "portfolio_value=1.004, cash_bias=0.2767, market_value=1.017, reward=-3.689e-06, episode=4943, loss=0.0002463\n", + "portfolio_value=1.026, cash_bias=0.2327, market_value=1.047, reward=-2.401e-05, episode=4959, loss=0.0002463\n", + "portfolio_value=0.9764, cash_bias=0.2318, market_value=1.013, reward=6.315e-06, episode=4975, loss=0.0002464\n", + "portfolio_value=1.037, cash_bias=0.1843, market_value=1.051, reward=-1.967e-05, episode=4991, loss=0.0002463\n", + "portfolio_value=1.015, cash_bias=0.1762, market_value=1.024, reward=-2.799e-05, episode=5007, loss=0.0002463\n", + "portfolio_value=1.034, cash_bias=0.3465, market_value=1.039, reward=-2.289e-05, episode=5023, loss=0.0002463\n", + "portfolio_value=1.049, cash_bias=0.1002, market_value=1.037, reward=1.723e-05, episode=5039, loss=0.0002462\n", + "portfolio_value=1.086, cash_bias=0.2444, market_value=1.079, reward=1.185e-06, episode=5055, loss=0.0002463\n", + "portfolio_value=1.062, cash_bias=0.2922, market_value=1.043, reward=-2.057e-05, episode=5071, loss=0.0002463\n", + "portfolio_value=1.028, cash_bias=0.2503, market_value=1.039, reward=-1.656e-05, episode=5087, loss=0.0002463\n", + "portfolio_value=1.025, cash_bias=0.2759, market_value=1.013, reward=2.174e-05, episode=5103, loss=0.0002463\n", + "portfolio_value=1.053, cash_bias=0.2402, market_value=1.022, reward=-5.179e-06, episode=5119, loss=0.0002462\n", + "portfolio_value=1.019, cash_bias=0.2487, market_value=1.038, reward=1.416e-05, episode=5135, loss=0.0002464\n", + "portfolio_value=1.019, cash_bias=0.127, market_value=1.017, reward=-1.56e-05, episode=5151, loss=0.0002464\n", + "portfolio_value=1.002, cash_bias=0.191, market_value=1.018, reward=-3.179e-05, episode=5167, loss=0.0002464\n", + "portfolio_value=1.045, cash_bias=0.1197, market_value=1.029, reward=-2.085e-05, episode=5183, loss=0.0002465\n", + "portfolio_value=1.05, cash_bias=0.2008, market_value=1.03, reward=1.955e-06, episode=5199, loss=0.0002463\n", + "portfolio_value=1.028, cash_bias=0.15, market_value=1.004, reward=1.295e-05, episode=5215, loss=0.0002463\n", + "portfolio_value=1.041, cash_bias=0.333, market_value=1.026, reward=9.579e-06, episode=5231, loss=0.0002462\n", + "portfolio_value=1.052, cash_bias=0.1898, market_value=1.045, reward=-1.966e-05, episode=5247, loss=0.0002463\n", + "portfolio_value=1.077, cash_bias=0.245, market_value=1.057, reward=3.082e-05, episode=5263, loss=0.0002463\n", + "portfolio_value=1.025, cash_bias=0.1741, market_value=1.027, reward=-1.176e-07, episode=5279, loss=0.0002463\n", + "portfolio_value=1.053, cash_bias=0.3096, market_value=1.031, reward=-7.734e-06, episode=5295, loss=0.0002463\n", + "portfolio_value=1.025, cash_bias=0.2226, market_value=1.017, reward=1.508e-05, episode=5311, loss=0.0002463\n", + "portfolio_value=1.057, cash_bias=0.1699, market_value=1.025, reward=-1.567e-05, episode=5327, loss=0.0002463\n", + "portfolio_value=1.028, cash_bias=0.1352, market_value=1.024, reward=8.877e-06, episode=5343, loss=0.0002463\n", + "portfolio_value=1.041, cash_bias=0.2458, market_value=1.04, reward=4.027e-06, episode=5359, loss=0.0002463\n", + "portfolio_value=1.053, cash_bias=0.09573, market_value=1.046, reward=1.876e-05, episode=5375, loss=0.0002463\n", + "portfolio_value=1.006, cash_bias=0.2841, market_value=1.019, reward=-2.242e-05, episode=5391, loss=0.0002463\n", + "portfolio_value=1.049, cash_bias=0.2, market_value=1.034, reward=2.485e-05, episode=5407, loss=0.0002465\n", + "portfolio_value=1.023, cash_bias=0.1747, market_value=1.005, reward=-9.894e-06, episode=5423, loss=0.0002464\n", + "portfolio_value=1.036, cash_bias=0.1936, market_value=1.051, reward=1.762e-05, episode=5439, loss=0.0002462\n", + "portfolio_value=1.022, cash_bias=0.2093, market_value=1.033, reward=1.803e-05, episode=5455, loss=0.0002463\n", + "portfolio_value=1.11, cash_bias=0.1504, market_value=1.068, reward=8.174e-06, episode=5471, loss=0.0002462\n", + "portfolio_value=1.039, cash_bias=0.2655, market_value=1.031, reward=-1.096e-05, episode=5487, loss=0.0002463\n", + "portfolio_value=1.04, cash_bias=0.2043, market_value=1.031, reward=-2.222e-05, episode=5503, loss=0.0002463\n", + "portfolio_value=1.005, cash_bias=0.2662, market_value=1.034, reward=-1.139e-05, episode=5519, loss=0.0002463\n", + "portfolio_value=1.026, cash_bias=0.1256, market_value=1.016, reward=6.468e-08, episode=5535, loss=0.0002463\n", + "portfolio_value=1.046, cash_bias=0.1659, market_value=1.061, reward=-2.088e-05, episode=5551, loss=0.0002462\n", + "portfolio_value=1.067, cash_bias=0.2492, market_value=1.061, reward=-3.142e-05, episode=5567, loss=0.0002463\n", + "portfolio_value=1.041, cash_bias=0.3096, market_value=1.038, reward=6.737e-06, episode=5583, loss=0.0002463\n", + "portfolio_value=1.038, cash_bias=0.2274, market_value=1.032, reward=-6.923e-06, episode=5599, loss=0.0002463\n", + "portfolio_value=1.103, cash_bias=0.3862, market_value=1.058, reward=-4.444e-05, episode=5615, loss=0.0002462\n", + "portfolio_value=1.021, cash_bias=0.2682, market_value=1.016, reward=-1.978e-05, episode=5631, loss=0.0002463\n", + "portfolio_value=1.022, cash_bias=0.185, market_value=1.028, reward=-1.447e-06, episode=5647, loss=0.0002463\n", + "portfolio_value=1.043, cash_bias=0.2661, market_value=1.057, reward=-9.898e-06, episode=5663, loss=0.0002463\n", + "portfolio_value=1.024, cash_bias=0.3182, market_value=1.06, reward=-5.43e-06, episode=5679, loss=0.0002463\n", + "portfolio_value=0.9873, cash_bias=0.2604, market_value=1.012, reward=-1.376e-05, episode=5695, loss=0.0002468\n", + "portfolio_value=1.024, cash_bias=0.1136, market_value=1.041, reward=-1.684e-05, episode=5711, loss=0.0002463\n", + "portfolio_value=1.049, cash_bias=0.1286, market_value=1.013, reward=-5.362e-06, episode=5727, loss=0.0002463\n", + "portfolio_value=1.037, cash_bias=0.2405, market_value=1.061, reward=1.292e-05, episode=5743, loss=0.0002463\n", + "portfolio_value=1.057, cash_bias=0.1751, market_value=1.009, reward=-4.915e-07, episode=5759, loss=0.0002464\n", + "portfolio_value=1.059, cash_bias=0.2473, market_value=1.046, reward=-4.188e-05, episode=5775, loss=0.0002462\n", + "portfolio_value=0.9936, cash_bias=0.1804, market_value=1.004, reward=2.903e-06, episode=5791, loss=0.0002464\n", + "portfolio_value=1.059, cash_bias=0.1684, market_value=1.032, reward=7.486e-06, episode=5807, loss=0.0002463\n", + "portfolio_value=1.051, cash_bias=0.2119, market_value=1.022, reward=-3.647e-06, episode=5823, loss=0.0002462\n", + "portfolio_value=1.074, cash_bias=0.3217, market_value=1.044, reward=-3.918e-06, episode=5839, loss=0.0002463\n", + "portfolio_value=1.048, cash_bias=0.2438, market_value=1.041, reward=5.917e-06, episode=5855, loss=0.0002462\n", + "portfolio_value=1.076, cash_bias=0.2116, market_value=1.06, reward=-2.802e-05, episode=5871, loss=0.0002463\n", + "portfolio_value=1.025, cash_bias=0.2458, market_value=1.026, reward=-1.197e-05, episode=5887, loss=0.0002463\n", + "portfolio_value=1.099, cash_bias=0.1201, market_value=1.045, reward=-1.476e-05, episode=5903, loss=0.0002462\n", + "portfolio_value=1.05, cash_bias=0.2705, market_value=1.035, reward=-6.3e-06, episode=5919, loss=0.0002462\n", + "portfolio_value=1.053, cash_bias=0.2456, market_value=1.049, reward=-1.478e-05, episode=5935, loss=0.0002463\n", + "portfolio_value=1.059, cash_bias=0.3752, market_value=1.023, reward=1.181e-05, episode=5951, loss=0.0002463\n", + "portfolio_value=1.034, cash_bias=0.2115, market_value=1.031, reward=-2.31e-06, episode=5967, loss=0.0002462\n", + "portfolio_value=1.039, cash_bias=0.3247, market_value=1.039, reward=-1.17e-05, episode=5983, loss=0.0002463\n", + "portfolio_value=1.007, cash_bias=0.3411, market_value=1.013, reward=1.073e-06, episode=5999, loss=0.0002463\n", + "portfolio_value=1.037, cash_bias=0.2928, market_value=1.038, reward=-6.082e-07, episode=6015, loss=0.0002463\n", + "portfolio_value=1.079, cash_bias=0.1602, market_value=1.056, reward=2.171e-05, episode=6031, loss=0.0002463\n", + "portfolio_value=1.069, cash_bias=0.1578, market_value=1.062, reward=-1.116e-05, episode=6047, loss=0.0002462\n", + "portfolio_value=1.026, cash_bias=0.1545, market_value=1.041, reward=-2.219e-05, episode=6063, loss=0.0002464\n", + "portfolio_value=1.018, cash_bias=0.3097, market_value=1.011, reward=6.922e-06, episode=6079, loss=0.0002463\n", + "portfolio_value=0.9966, cash_bias=0.3194, market_value=1.02, reward=1.908e-05, episode=6095, loss=0.0002463\n", + "portfolio_value=1.011, cash_bias=0.2701, market_value=1.021, reward=-2.51e-05, episode=6111, loss=0.0002463\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.032, cash_bias=0.3458, market_value=1.06, reward=-3.472e-05, episode=6127, loss=0.0002463\n", + "portfolio_value=1.021, cash_bias=0.2263, market_value=1.031, reward=3.84e-05, episode=6143, loss=0.0002463\n", + "portfolio_value=1.058, cash_bias=0.2469, market_value=1.068, reward=8.301e-06, episode=6159, loss=0.0002463\n", + "portfolio_value=1.021, cash_bias=0.3243, market_value=1.029, reward=-1.039e-05, episode=6175, loss=0.0002463\n", + "portfolio_value=1.054, cash_bias=0.2013, market_value=1.023, reward=-2.087e-05, episode=6191, loss=0.0002463\n", + "portfolio_value=1.067, cash_bias=0.3865, market_value=1.043, reward=1.301e-05, episode=6207, loss=0.0002462\n", + "portfolio_value=1.064, cash_bias=0.3071, market_value=1.065, reward=1.595e-06, episode=6223, loss=0.0002462\n", + "portfolio_value=1.011, cash_bias=0.2504, market_value=1.029, reward=-1.24e-05, episode=6239, loss=0.0002463\n", + "portfolio_value=1.021, cash_bias=0.1794, market_value=1.014, reward=-8.65e-06, episode=6255, loss=0.0002463\n", + "portfolio_value=1.027, cash_bias=0.1193, market_value=1.034, reward=5.781e-06, episode=6271, loss=0.0002463\n", + "portfolio_value=1.028, cash_bias=0.26, market_value=1.015, reward=-2.103e-05, episode=6287, loss=0.0002463\n", + "portfolio_value=1.076, cash_bias=0.1438, market_value=1.051, reward=3.623e-05, episode=6303, loss=0.0002462\n", + "portfolio_value=1.069, cash_bias=0.4465, market_value=1.067, reward=-1.066e-05, episode=6319, loss=0.0002462\n", + "portfolio_value=1.029, cash_bias=0.2047, market_value=1.008, reward=-5.067e-06, episode=6335, loss=0.0002463\n", + "portfolio_value=1.058, cash_bias=0.2863, market_value=1.037, reward=-5.953e-06, episode=6351, loss=0.0002462\n", + "portfolio_value=1.042, cash_bias=0.4007, market_value=1.04, reward=-4.002e-06, episode=6367, loss=0.0002463\n", + "portfolio_value=1.015, cash_bias=0.288, market_value=1.026, reward=2.228e-05, episode=6383, loss=0.0002463\n", + "portfolio_value=1.058, cash_bias=0.1637, market_value=1.032, reward=5.346e-06, episode=6399, loss=0.0002462\n", + "portfolio_value=1.002, cash_bias=0.2686, market_value=0.9964, reward=3.224e-05, episode=6415, loss=0.0002463\n", + "portfolio_value=1.056, cash_bias=0.1999, market_value=1.071, reward=5.906e-06, episode=6431, loss=0.0002463\n", + "portfolio_value=1.085, cash_bias=0.2014, market_value=1.052, reward=2.186e-05, episode=6447, loss=0.0002462\n", + "portfolio_value=1.039, cash_bias=0.2325, market_value=1.035, reward=1.944e-06, episode=6463, loss=0.0002462\n", + "portfolio_value=1.021, cash_bias=0.1315, market_value=1.03, reward=2.329e-05, episode=6479, loss=0.0002463\n", + "portfolio_value=1.044, cash_bias=0.1874, market_value=1.035, reward=-9.314e-06, episode=6495, loss=0.0002463\n", + "portfolio_value=0.9967, cash_bias=0.1097, market_value=1.015, reward=-8.927e-06, episode=6511, loss=0.0002463\n", + "portfolio_value=1.011, cash_bias=0.2924, market_value=1.036, reward=1.743e-05, episode=6527, loss=0.0002463\n", + "portfolio_value=1.042, cash_bias=0.1511, market_value=1.045, reward=-1.094e-05, episode=6543, loss=0.0002463\n", + "portfolio_value=1.025, cash_bias=0.1159, market_value=1.021, reward=3.498e-06, episode=6559, loss=0.0002463\n", + "portfolio_value=1.028, cash_bias=0.2573, market_value=1.035, reward=4.315e-06, episode=6575, loss=0.0002463\n", + "portfolio_value=1.032, cash_bias=0.1595, market_value=1.032, reward=-2.163e-06, episode=6591, loss=0.0002463\n", + "portfolio_value=1.01, cash_bias=0.1842, market_value=1.026, reward=1.614e-05, episode=6607, loss=0.0002463\n", + "portfolio_value=1.021, cash_bias=0.3038, market_value=1.033, reward=1.644e-06, episode=6623, loss=0.0002464\n", + "portfolio_value=1.032, cash_bias=0.2567, market_value=1.032, reward=1.425e-05, episode=6639, loss=0.0002463\n", + "portfolio_value=1.054, cash_bias=0.2134, market_value=1.057, reward=9.075e-06, episode=6655, loss=0.0002463\n", + "portfolio_value=1.018, cash_bias=0.2572, market_value=1.042, reward=-9.601e-06, episode=6671, loss=0.0002463\n", + "portfolio_value=1.005, cash_bias=0.2412, market_value=0.9967, reward=1.822e-05, episode=6687, loss=0.0002464\n", + "portfolio_value=0.9899, cash_bias=0.2988, market_value=1.005, reward=-1.885e-05, episode=6703, loss=0.0002463\n", + "portfolio_value=0.9993, cash_bias=0.1345, market_value=1.035, reward=-6.096e-06, episode=6719, loss=0.0002463\n", + "portfolio_value=1.064, cash_bias=0.1513, market_value=1.043, reward=-1.583e-05, episode=6735, loss=0.0002462\n", + "portfolio_value=1.055, cash_bias=0.1738, market_value=1.048, reward=6.16e-07, episode=6751, loss=0.0002463\n", + "portfolio_value=1.023, cash_bias=0.3163, market_value=1.033, reward=7.376e-06, episode=6767, loss=0.0002463\n", + "portfolio_value=0.9953, cash_bias=0.2347, market_value=1.007, reward=-8.598e-07, episode=6783, loss=0.0002464\n", + "portfolio_value=1.034, cash_bias=0.1766, market_value=1.02, reward=-2.497e-06, episode=6799, loss=0.0002464\n", + "portfolio_value=1.019, cash_bias=0.1447, market_value=1.021, reward=3.231e-05, episode=6815, loss=0.0002463\n", + "portfolio_value=1.026, cash_bias=0.1749, market_value=1.024, reward=4.877e-07, episode=6831, loss=0.0002463\n", + "portfolio_value=1.05, cash_bias=0.1504, market_value=1.039, reward=1.953e-05, episode=6847, loss=0.0002463\n", + "portfolio_value=1.052, cash_bias=0.3142, market_value=1.051, reward=3.98e-05, episode=6863, loss=0.0002463\n", + "portfolio_value=1.056, cash_bias=0.2829, market_value=1.057, reward=1.621e-05, episode=6879, loss=0.0002463\n", + "portfolio_value=1.029, cash_bias=0.1619, market_value=1.028, reward=1.427e-05, episode=6895, loss=0.0002463\n", + "portfolio_value=1.024, cash_bias=0.2566, market_value=1.036, reward=1.285e-05, episode=6911, loss=0.0002463\n", + "portfolio_value=1.048, cash_bias=0.2325, market_value=1.05, reward=3.54e-05, episode=6927, loss=0.0002463\n", + "portfolio_value=1.028, cash_bias=0.1002, market_value=1.057, reward=-7.384e-06, episode=6943, loss=0.0002463\n", + "portfolio_value=1.048, cash_bias=0.3046, market_value=1.028, reward=-6.047e-05, episode=6959, loss=0.0002463\n", + "portfolio_value=1.041, cash_bias=0.2748, market_value=1.038, reward=-8.502e-06, episode=6975, loss=0.0002463\n", + "portfolio_value=0.988, cash_bias=0.3842, market_value=1.016, reward=3.296e-05, episode=6991, loss=0.0002464\n", + "portfolio_value=1.055, cash_bias=0.1451, market_value=1.035, reward=-2.207e-06, episode=7007, loss=0.0002463\n", + "portfolio_value=1.011, cash_bias=0.1351, market_value=1.023, reward=5.111e-06, episode=7023, loss=0.0002463\n", + "portfolio_value=1.065, cash_bias=0.1331, market_value=1.033, reward=6.707e-06, episode=7039, loss=0.0002462\n", + "portfolio_value=1.043, cash_bias=0.3225, market_value=1.034, reward=-1.011e-05, episode=7055, loss=0.0002463\n", + "portfolio_value=1.054, cash_bias=0.3228, market_value=1.042, reward=-5.135e-06, episode=7071, loss=0.0002463\n", + "portfolio_value=1.069, cash_bias=0.2617, market_value=1.056, reward=-2.308e-05, episode=7087, loss=0.0002463\n", + "portfolio_value=0.9952, cash_bias=0.186, market_value=1.037, reward=-2.106e-05, episode=7103, loss=0.0002464\n", + "portfolio_value=1.081, cash_bias=0.3128, market_value=1.055, reward=-4.157e-06, episode=7119, loss=0.0002462\n", + "portfolio_value=1.058, cash_bias=0.2591, market_value=1.054, reward=4.443e-05, episode=7135, loss=0.0002463\n", + "portfolio_value=1.058, cash_bias=0.3064, market_value=1.065, reward=1.241e-05, episode=7151, loss=0.0002463\n", + "portfolio_value=1.011, cash_bias=0.2663, market_value=1.024, reward=-2.277e-06, episode=7167, loss=0.0002463\n", + "portfolio_value=1.036, cash_bias=0.3605, market_value=1.037, reward=-2.138e-05, episode=7183, loss=0.0002463\n", + "portfolio_value=1.052, cash_bias=0.2187, market_value=1.06, reward=1.334e-05, episode=7199, loss=0.0002463\n", + "portfolio_value=1.048, cash_bias=0.2402, market_value=1.004, reward=8.621e-06, episode=7215, loss=0.0002463\n", + "portfolio_value=1.067, cash_bias=0.2877, market_value=1.059, reward=-4.399e-05, episode=7231, loss=0.0002462\n", + "portfolio_value=1.01, cash_bias=0.1778, market_value=1.025, reward=-2.36e-05, episode=7247, loss=0.0002463\n", + "portfolio_value=1.042, cash_bias=0.1313, market_value=1.028, reward=-1.199e-05, episode=7263, loss=0.0002463\n", + "portfolio_value=1.047, cash_bias=0.2435, market_value=1.029, reward=1.878e-05, episode=7279, loss=0.0002468\n", + "portfolio_value=1.051, cash_bias=0.2061, market_value=1.035, reward=8.468e-06, episode=7295, loss=0.0002463\n", + "portfolio_value=1.067, cash_bias=0.4552, market_value=1.043, reward=8.591e-06, episode=7311, loss=0.0002462\n", + "portfolio_value=1.062, cash_bias=0.1956, market_value=1.06, reward=-3.443e-06, episode=7327, loss=0.0002462\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.008, cash_bias=0.3215, market_value=1.009, reward=-4.55e-06, episode=7343, loss=0.0002463\n", + "portfolio_value=1.057, cash_bias=0.3237, market_value=1.061, reward=1.097e-05, episode=7359, loss=0.0002463\n", + "portfolio_value=1.086, cash_bias=0.3549, market_value=1.053, reward=-4.186e-06, episode=7375, loss=0.0002463\n", + "portfolio_value=1.054, cash_bias=0.2391, market_value=1.027, reward=-1.189e-05, episode=7391, loss=0.0002462\n", + "portfolio_value=1.024, cash_bias=0.1439, market_value=1.022, reward=-2.281e-05, episode=7407, loss=0.0002463\n", + "portfolio_value=1.033, cash_bias=0.2965, market_value=1.042, reward=1.282e-05, episode=7423, loss=0.0002463\n", + "portfolio_value=1.048, cash_bias=0.2399, market_value=1.018, reward=1.411e-05, episode=7439, loss=0.0002463\n", + "portfolio_value=1.041, cash_bias=0.3339, market_value=1.038, reward=1.768e-06, episode=7455, loss=0.0002462\n", + "portfolio_value=1.057, cash_bias=0.2004, market_value=1.043, reward=-3.453e-05, episode=7471, loss=0.0002463\n", + "portfolio_value=1.017, cash_bias=0.2266, market_value=1.015, reward=3.814e-06, episode=7487, loss=0.0002463\n", + "portfolio_value=1.044, cash_bias=0.2008, market_value=1.032, reward=7.304e-06, episode=7503, loss=0.0002462\n", + "portfolio_value=1.034, cash_bias=0.2345, market_value=1.035, reward=3.597e-06, episode=7519, loss=0.0002463\n", + "portfolio_value=0.9794, cash_bias=0.1819, market_value=1.011, reward=-1.035e-06, episode=7535, loss=0.0002467\n", + "portfolio_value=1.052, cash_bias=0.1833, market_value=1.055, reward=3.087e-06, episode=7551, loss=0.0002463\n", + "portfolio_value=1.047, cash_bias=0.2112, market_value=1.032, reward=2.136e-05, episode=7567, loss=0.0002462\n", + "portfolio_value=1.047, cash_bias=0.3415, market_value=1.009, reward=-1.582e-05, episode=7583, loss=0.0002462\n", + "portfolio_value=0.9958, cash_bias=0.2094, market_value=1.018, reward=2.707e-05, episode=7599, loss=0.0002463\n", + "portfolio_value=1.031, cash_bias=0.1533, market_value=1.031, reward=-1.794e-05, episode=7615, loss=0.0002463\n", + "portfolio_value=1.073, cash_bias=0.2553, market_value=1.067, reward=7.809e-06, episode=7631, loss=0.0002463\n", + "portfolio_value=1.065, cash_bias=0.3099, market_value=1.056, reward=8.262e-06, episode=7647, loss=0.0002463\n", + "portfolio_value=1.07, cash_bias=0.269, market_value=1.061, reward=2.006e-05, episode=7663, loss=0.0002462\n", + "portfolio_value=1.011, cash_bias=0.238, market_value=1.034, reward=4.028e-05, episode=7679, loss=0.0002464\n", + "portfolio_value=1.09, cash_bias=0.3091, market_value=1.028, reward=-1.31e-05, episode=7695, loss=0.0002462\n", + "portfolio_value=0.992, cash_bias=0.189, market_value=0.9953, reward=7.043e-07, episode=7711, loss=0.0002468\n", + "portfolio_value=1.002, cash_bias=0.1906, market_value=1.008, reward=3.987e-05, episode=7727, loss=0.0002468\n", + "portfolio_value=1.05, cash_bias=0.1973, market_value=1.039, reward=2.872e-05, episode=7743, loss=0.0002463\n", + "portfolio_value=1.086, cash_bias=0.1539, market_value=1.055, reward=3.352e-05, episode=7759, loss=0.0002462\n", + "portfolio_value=1.065, cash_bias=0.1999, market_value=1.055, reward=1.819e-05, episode=7775, loss=0.0002462\n", + "portfolio_value=1.009, cash_bias=0.4336, market_value=1.029, reward=2.338e-05, episode=7791, loss=0.0002463\n", + "portfolio_value=1.062, cash_bias=0.2801, market_value=1.064, reward=8.291e-07, episode=7807, loss=0.0002462\n", + "portfolio_value=1.017, cash_bias=0.1569, market_value=1.013, reward=-1.274e-05, episode=7823, loss=0.000247\n", + "portfolio_value=1.025, cash_bias=0.1951, market_value=1.024, reward=-2.771e-06, episode=7839, loss=0.0002464\n", + "portfolio_value=1.044, cash_bias=0.2947, market_value=1.038, reward=-1.028e-05, episode=7855, loss=0.0002463\n", + "portfolio_value=1.06, cash_bias=0.145, market_value=1.034, reward=1.311e-05, episode=7871, loss=0.0002462\n", + "portfolio_value=1.065, cash_bias=0.1922, market_value=1.037, reward=3.287e-05, episode=7887, loss=0.0002463\n", + "portfolio_value=1.066, cash_bias=0.2869, market_value=1.036, reward=-4.386e-06, episode=7903, loss=0.0002463\n", + "portfolio_value=1.032, cash_bias=0.1484, market_value=1.02, reward=-3.461e-06, episode=7919, loss=0.0002463\n", + "portfolio_value=1.023, cash_bias=0.2427, market_value=1.038, reward=-4.211e-06, episode=7935, loss=0.0002463\n", + "portfolio_value=1.025, cash_bias=0.2736, market_value=1.052, reward=-1.778e-05, episode=7951, loss=0.0002463\n", + "portfolio_value=1.077, cash_bias=0.3138, market_value=1.075, reward=4.37e-06, episode=7967, loss=0.0002462\n", + "portfolio_value=1.08, cash_bias=0.1762, market_value=1.061, reward=-5.108e-06, episode=7983, loss=0.0002462\n", + "portfolio_value=1.045, cash_bias=0.2212, market_value=1.039, reward=-6.805e-06, episode=7999, loss=0.0002463\n", + "portfolio_value=1.055, cash_bias=0.3416, market_value=1.045, reward=4.497e-05, episode=8015, loss=0.0002463\n", + "portfolio_value=1.052, cash_bias=0.1064, market_value=1.045, reward=-1.115e-05, episode=8031, loss=0.0002463\n", + "portfolio_value=1.05, cash_bias=0.1989, market_value=1.039, reward=6.442e-05, episode=8047, loss=0.0002463\n", + "portfolio_value=0.9884, cash_bias=0.1367, market_value=1.019, reward=-5.097e-06, episode=8063, loss=0.0002464\n", + "portfolio_value=1.033, cash_bias=0.2692, market_value=1.036, reward=1.366e-05, episode=8079, loss=0.0002463\n", + "portfolio_value=1.03, cash_bias=0.1829, market_value=1.038, reward=2.063e-05, episode=8095, loss=0.0002463\n", + "portfolio_value=1.079, cash_bias=0.29, market_value=1.047, reward=1.988e-05, episode=8111, loss=0.0002462\n", + "portfolio_value=1.057, cash_bias=0.1327, market_value=1.073, reward=-3.064e-06, episode=8127, loss=0.0002463\n", + "portfolio_value=1.037, cash_bias=0.2016, market_value=1.035, reward=1.15e-05, episode=8143, loss=0.0002463\n", + "portfolio_value=1.013, cash_bias=0.1696, market_value=1.015, reward=-1.143e-05, episode=8159, loss=0.0002464\n", + "portfolio_value=1.09, cash_bias=0.244, market_value=1.064, reward=9.135e-06, episode=8175, loss=0.0002462\n", + "portfolio_value=1.038, cash_bias=0.1811, market_value=1.048, reward=1.81e-05, episode=8191, loss=0.0002463\n", + "portfolio_value=1.059, cash_bias=0.3286, market_value=1.045, reward=2.539e-06, episode=8207, loss=0.0002462\n", + "portfolio_value=1.037, cash_bias=0.2154, market_value=1.068, reward=3.424e-06, episode=8223, loss=0.0002463\n", + "portfolio_value=1.049, cash_bias=0.3105, market_value=1.014, reward=1.07e-05, episode=8239, loss=0.0002463\n", + "portfolio_value=1.026, cash_bias=0.2592, market_value=1.044, reward=3.568e-06, episode=8255, loss=0.0002464\n", + "portfolio_value=1.009, cash_bias=0.319, market_value=1.044, reward=-1.784e-05, episode=8271, loss=0.0002463\n", + "portfolio_value=1.041, cash_bias=0.3014, market_value=1.012, reward=-7.198e-06, episode=8287, loss=0.0002463\n", + "portfolio_value=1.071, cash_bias=0.2399, market_value=1.038, reward=4.658e-05, episode=8303, loss=0.0002463\n", + "portfolio_value=1.045, cash_bias=0.1357, market_value=1.028, reward=-1.017e-05, episode=8319, loss=0.0002463\n", + "portfolio_value=1.007, cash_bias=0.35, market_value=1.016, reward=3.107e-06, episode=8335, loss=0.0002463\n", + "portfolio_value=1.097, cash_bias=0.1704, market_value=1.072, reward=-2.136e-05, episode=8351, loss=0.0002462\n", + "portfolio_value=1.052, cash_bias=0.2753, market_value=1.048, reward=3.459e-05, episode=8367, loss=0.0002463\n", + "portfolio_value=1.063, cash_bias=0.1587, market_value=1.018, reward=-1.54e-06, episode=8383, loss=0.0002462\n", + "portfolio_value=0.9788, cash_bias=0.1934, market_value=1.013, reward=4.331e-06, episode=8399, loss=0.0002464\n", + "portfolio_value=1.028, cash_bias=0.2583, market_value=1.015, reward=3.533e-05, episode=8415, loss=0.0002463\n", + "portfolio_value=1.087, cash_bias=0.2851, market_value=1.047, reward=1.516e-06, episode=8431, loss=0.0002462\n", + "portfolio_value=1.022, cash_bias=0.226, market_value=1.039, reward=-2.839e-06, episode=8447, loss=0.0002463\n", + "portfolio_value=1.046, cash_bias=0.1795, market_value=1.012, reward=-1.008e-05, episode=8463, loss=0.0002463\n", + "portfolio_value=1.047, cash_bias=0.1918, market_value=1.017, reward=-2.374e-05, episode=8479, loss=0.0002463\n", + "portfolio_value=1.067, cash_bias=0.1837, market_value=1.053, reward=-4.439e-06, episode=8495, loss=0.0002462\n", + "portfolio_value=1.051, cash_bias=0.1419, market_value=1.025, reward=5.041e-06, episode=8511, loss=0.0002463\n", + "portfolio_value=1.036, cash_bias=0.2783, market_value=1.048, reward=-1.295e-05, episode=8527, loss=0.0002462\n", + "portfolio_value=1.022, cash_bias=0.1105, market_value=1.019, reward=-1.917e-05, episode=8543, loss=0.0002463\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.009, cash_bias=0.2558, market_value=1.016, reward=-4.616e-06, episode=8559, loss=0.0002463\n", + "portfolio_value=1.049, cash_bias=0.2429, market_value=1.032, reward=1.861e-06, episode=8575, loss=0.0002462\n", + "portfolio_value=1.026, cash_bias=0.1643, market_value=1.055, reward=-6.888e-06, episode=8591, loss=0.0002463\n", + "portfolio_value=1.053, cash_bias=0.2668, market_value=1.035, reward=-2.599e-05, episode=8607, loss=0.0002463\n", + "portfolio_value=0.9998, cash_bias=0.257, market_value=1.022, reward=6.93e-06, episode=8623, loss=0.0002464\n", + "portfolio_value=1.017, cash_bias=0.3561, market_value=1.039, reward=-1.859e-05, episode=8639, loss=0.0002463\n", + "portfolio_value=1.014, cash_bias=0.2465, market_value=1.015, reward=-8.17e-06, episode=8655, loss=0.0002463\n", + "portfolio_value=1.006, cash_bias=0.2388, market_value=1.006, reward=-1.792e-05, episode=8671, loss=0.0002463\n", + "portfolio_value=1.065, cash_bias=0.1173, market_value=1.031, reward=3.14e-07, episode=8687, loss=0.0002462\n", + "portfolio_value=1.057, cash_bias=0.2705, market_value=1.043, reward=2.715e-05, episode=8703, loss=0.0002462\n", + "portfolio_value=1.022, cash_bias=0.1708, market_value=1.023, reward=1.796e-05, episode=8719, loss=0.0002463\n", + "portfolio_value=0.9824, cash_bias=0.2116, market_value=0.9892, reward=-1.748e-05, episode=8735, loss=0.0002463\n", + "portfolio_value=1.023, cash_bias=0.2187, market_value=1.032, reward=5.062e-06, episode=8751, loss=0.0002463\n", + "portfolio_value=1.088, cash_bias=0.2225, market_value=1.057, reward=-5.563e-06, episode=8767, loss=0.0002462\n", + "portfolio_value=1.031, cash_bias=0.2596, market_value=1.024, reward=1.432e-05, episode=8783, loss=0.0002463\n", + "portfolio_value=1.034, cash_bias=0.2575, market_value=1.044, reward=4.024e-06, episode=8799, loss=0.0002463\n", + "portfolio_value=1.04, cash_bias=0.1678, market_value=1.03, reward=9.113e-06, episode=8815, loss=0.0002462\n", + "portfolio_value=1.016, cash_bias=0.1472, market_value=1.01, reward=-1.406e-05, episode=8831, loss=0.0002463\n", + "portfolio_value=1.075, cash_bias=0.2604, market_value=1.045, reward=1.413e-05, episode=8847, loss=0.0002462\n", + "portfolio_value=1.01, cash_bias=0.202, market_value=1.007, reward=-9.758e-06, episode=8863, loss=0.0002464\n", + "portfolio_value=1.037, cash_bias=0.1322, market_value=1.041, reward=-3.349e-05, episode=8879, loss=0.0002462\n", + "portfolio_value=1.026, cash_bias=0.2816, market_value=1.027, reward=6.37e-06, episode=8895, loss=0.0002463\n", + "portfolio_value=1.037, cash_bias=0.2681, market_value=1.034, reward=8.037e-06, episode=8911, loss=0.0002463\n", + "portfolio_value=1.017, cash_bias=0.2865, market_value=1.027, reward=3.51e-05, episode=8927, loss=0.0002463\n", + "portfolio_value=1.045, cash_bias=0.19, market_value=1.054, reward=1.088e-05, episode=8943, loss=0.0002462\n", + "portfolio_value=1.03, cash_bias=0.06688, market_value=1.026, reward=-1.436e-05, episode=8959, loss=0.0002463\n", + "portfolio_value=1.074, cash_bias=0.2708, market_value=1.04, reward=-3.022e-06, episode=8975, loss=0.0002463\n", + "portfolio_value=1.047, cash_bias=0.1855, market_value=1.04, reward=-2.365e-05, episode=8991, loss=0.0002463\n", + "portfolio_value=1.08, cash_bias=0.1246, market_value=1.046, reward=1.243e-05, episode=9007, loss=0.0002462\n", + "portfolio_value=1.051, cash_bias=0.2004, market_value=1.021, reward=1.005e-05, episode=9023, loss=0.0002463\n", + "portfolio_value=1.068, cash_bias=0.2501, market_value=1.046, reward=2.59e-05, episode=9039, loss=0.0002463\n", + "portfolio_value=1.042, cash_bias=0.3001, market_value=1.035, reward=1.69e-05, episode=9055, loss=0.0002463\n", + "portfolio_value=1.049, cash_bias=0.2215, market_value=1.03, reward=4.526e-06, episode=9071, loss=0.0002462\n", + "portfolio_value=1.052, cash_bias=0.1573, market_value=1.013, reward=-2.166e-05, episode=9087, loss=0.0002468\n", + "portfolio_value=1.031, cash_bias=0.2591, market_value=1.023, reward=1.446e-05, episode=9103, loss=0.0002472\n", + "portfolio_value=1.025, cash_bias=0.372, market_value=1.022, reward=2.632e-07, episode=9119, loss=0.0002477\n", + "portfolio_value=1.04, cash_bias=0.1986, market_value=1.03, reward=3.165e-05, episode=9135, loss=0.000248\n", + "portfolio_value=1.048, cash_bias=0.2182, market_value=1.024, reward=7.232e-06, episode=9151, loss=0.000248\n", + "portfolio_value=1.063, cash_bias=0.2991, market_value=1.037, reward=-1.814e-05, episode=9167, loss=0.0002481\n", + "portfolio_value=1.054, cash_bias=0.3335, market_value=1.03, reward=2.76e-05, episode=9183, loss=0.000248\n", + "portfolio_value=1.055, cash_bias=0.2531, market_value=1.033, reward=-1.541e-05, episode=9199, loss=0.000248\n", + "portfolio_value=1.003, cash_bias=0.1544, market_value=1.007, reward=9.929e-06, episode=9215, loss=0.0002481\n", + "portfolio_value=1.048, cash_bias=0.1318, market_value=1.044, reward=-1.142e-05, episode=9231, loss=0.000248\n", + "portfolio_value=1.066, cash_bias=0.2003, market_value=1.047, reward=1.282e-05, episode=9247, loss=0.000248\n", + "portfolio_value=0.9946, cash_bias=0.2764, market_value=1.013, reward=-2.01e-05, episode=9263, loss=0.0002481\n", + "portfolio_value=1.037, cash_bias=0.2779, market_value=1.03, reward=-5.572e-06, episode=9279, loss=0.000248\n", + "portfolio_value=1.015, cash_bias=0.1569, market_value=1.036, reward=1.891e-05, episode=9295, loss=0.0002481\n", + "portfolio_value=1.012, cash_bias=0.2091, market_value=1.018, reward=5.919e-06, episode=9311, loss=0.0002481\n", + "portfolio_value=1.042, cash_bias=0.2895, market_value=1.046, reward=-2.443e-05, episode=9327, loss=0.0002481\n", + "portfolio_value=1.007, cash_bias=0.1242, market_value=1.03, reward=7.985e-06, episode=9343, loss=0.0002481\n", + "portfolio_value=1.065, cash_bias=0.2182, market_value=1.048, reward=-7.488e-06, episode=9359, loss=0.000248\n", + "portfolio_value=0.9856, cash_bias=0.1935, market_value=1.001, reward=8.21e-06, episode=9375, loss=0.0002481\n", + "portfolio_value=1.062, cash_bias=0.2063, market_value=1.022, reward=6.035e-06, episode=9391, loss=0.000248\n", + "portfolio_value=1.027, cash_bias=0.1415, market_value=1.02, reward=3.092e-06, episode=9407, loss=0.0002481\n", + "portfolio_value=1.021, cash_bias=0.2026, market_value=1.043, reward=2.274e-05, episode=9423, loss=0.0002481\n", + "portfolio_value=0.9966, cash_bias=0.1421, market_value=1.025, reward=2.052e-05, episode=9439, loss=0.0002482\n", + "portfolio_value=1.042, cash_bias=0.289, market_value=1.02, reward=3.547e-06, episode=9455, loss=0.000248\n", + "portfolio_value=1.047, cash_bias=0.3324, market_value=1.013, reward=-1.302e-05, episode=9471, loss=0.000248\n", + "portfolio_value=1.04, cash_bias=0.252, market_value=1.034, reward=-4.647e-06, episode=9487, loss=0.0002481\n", + "portfolio_value=0.9968, cash_bias=0.1665, market_value=1.034, reward=-1.789e-05, episode=9503, loss=0.0002481\n", + "portfolio_value=1.066, cash_bias=0.2487, market_value=1.067, reward=-1.637e-05, episode=9519, loss=0.000248\n", + "portfolio_value=0.9978, cash_bias=0.2431, market_value=1.027, reward=-4.396e-06, episode=9535, loss=0.0002481\n", + "portfolio_value=1.067, cash_bias=0.2723, market_value=1.061, reward=-1.244e-05, episode=9551, loss=0.000248\n", + "portfolio_value=1.039, cash_bias=0.3491, market_value=1.063, reward=1.088e-05, episode=9567, loss=0.000248\n", + "portfolio_value=1.027, cash_bias=0.3376, market_value=1.01, reward=5.532e-07, episode=9583, loss=0.000248\n", + "portfolio_value=1.065, cash_bias=0.1722, market_value=1.047, reward=1.184e-05, episode=9599, loss=0.000248\n", + "portfolio_value=1.029, cash_bias=0.1674, market_value=1.023, reward=1.089e-05, episode=9615, loss=0.000248\n", + "portfolio_value=1.121, cash_bias=0.2001, market_value=1.072, reward=7.683e-06, episode=9631, loss=0.000248\n", + "portfolio_value=1.012, cash_bias=0.1711, market_value=1.018, reward=5.757e-06, episode=9647, loss=0.0002481\n", + "portfolio_value=1.088, cash_bias=0.2449, market_value=1.056, reward=-2.193e-05, episode=9663, loss=0.000248\n", + "portfolio_value=1.034, cash_bias=0.3109, market_value=1.026, reward=1.035e-05, episode=9679, loss=0.0002481\n", + "portfolio_value=1.004, cash_bias=0.3391, market_value=1.029, reward=3.646e-05, episode=9695, loss=0.0002487\n", + "portfolio_value=1.048, cash_bias=0.1514, market_value=1.021, reward=1.022e-05, episode=9711, loss=0.0002481\n", + "portfolio_value=1.063, cash_bias=0.2403, market_value=1.047, reward=4.613e-06, episode=9727, loss=0.000248\n", + "portfolio_value=1.057, cash_bias=0.1615, market_value=1.048, reward=9.803e-06, episode=9743, loss=0.000248\n", + "portfolio_value=1.048, cash_bias=0.2648, market_value=1.03, reward=1.689e-05, episode=9759, loss=0.000248\n", + "portfolio_value=1.026, cash_bias=0.4004, market_value=1.028, reward=2.174e-05, episode=9775, loss=0.0002481\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.011, cash_bias=0.2272, market_value=1.025, reward=1.072e-05, episode=9791, loss=0.0002481\n", + "portfolio_value=1.1, cash_bias=0.2078, market_value=1.067, reward=-9.594e-06, episode=9807, loss=0.000248\n", + "portfolio_value=1.049, cash_bias=0.1428, market_value=1.056, reward=2.568e-05, episode=9823, loss=0.000248\n", + "portfolio_value=1.004, cash_bias=0.3183, market_value=1.01, reward=1.008e-05, episode=9839, loss=0.0002481\n", + "portfolio_value=1.037, cash_bias=0.1731, market_value=1.034, reward=3.989e-05, episode=9855, loss=0.000248\n", + "portfolio_value=1.066, cash_bias=0.2188, market_value=1.043, reward=1.396e-05, episode=9871, loss=0.000248\n", + "portfolio_value=1.006, cash_bias=0.2059, market_value=1.026, reward=-1.016e-05, episode=9887, loss=0.0002481\n", + "portfolio_value=1.036, cash_bias=0.1549, market_value=1.028, reward=1.749e-05, episode=9903, loss=0.000248\n", + "portfolio_value=1.013, cash_bias=0.209, market_value=1.036, reward=1.165e-05, episode=9919, loss=0.0002481\n", + "portfolio_value=1.003, cash_bias=0.2268, market_value=1.012, reward=-1.09e-05, episode=9935, loss=0.0002483\n", + "portfolio_value=1.033, cash_bias=0.175, market_value=1.058, reward=-3.638e-05, episode=9951, loss=0.000248\n", + "portfolio_value=1.009, cash_bias=0.3131, market_value=1.019, reward=2.91e-05, episode=9967, loss=0.0002481\n", + "portfolio_value=1.044, cash_bias=0.2147, market_value=1.025, reward=-8.904e-06, episode=9983, loss=0.000248\n", + "portfolio_value=1.031, cash_bias=0.1605, market_value=1.034, reward=1.779e-06, episode=9999, loss=0.0002485\n", + "portfolio_value=1.062, cash_bias=0.2998, market_value=1.053, reward=-3.639e-06, episode=1.002e+04, loss=0.000248\n", + "portfolio_value=1.084, cash_bias=0.2214, market_value=1.058, reward=-1.74e-05, episode=1.003e+04, loss=0.000248\n", + "portfolio_value=0.977, cash_bias=0.4001, market_value=1.01, reward=8.382e-06, episode=1.005e+04, loss=0.0002481\n", + "portfolio_value=1.029, cash_bias=0.2341, market_value=1.018, reward=1.164e-06, episode=1.006e+04, loss=0.000248\n", + "portfolio_value=1.052, cash_bias=0.3064, market_value=1.045, reward=1.683e-05, episode=1.008e+04, loss=0.000248\n", + "portfolio_value=1.041, cash_bias=0.21, market_value=1.028, reward=3.098e-06, episode=1.01e+04, loss=0.000248\n", + "portfolio_value=1.006, cash_bias=0.2497, market_value=1.029, reward=1.038e-05, episode=1.011e+04, loss=0.0002481\n", + "portfolio_value=1.063, cash_bias=0.1353, market_value=1.051, reward=2.218e-05, episode=1.013e+04, loss=0.000248\n", + "portfolio_value= 1, cash_bias=0.1812, market_value=1.026, reward=1.366e-05, episode=1.014e+04, loss=0.0002481\n", + "portfolio_value=1.047, cash_bias=0.3866, market_value=1.034, reward=-7.355e-06, episode=1.016e+04, loss=0.0002481\n", + "portfolio_value= 1, cash_bias=0.1808, market_value=1.027, reward=-1.483e-05, episode=1.018e+04, loss=0.0002481\n", + "portfolio_value=1.056, cash_bias=0.1619, market_value=1.058, reward=-1.34e-07, episode=1.019e+04, loss=0.000248\n", + "portfolio_value=1.08, cash_bias=0.2146, market_value=1.033, reward=-1.551e-05, episode=1.021e+04, loss=0.000248\n", + "portfolio_value=1.053, cash_bias=0.2379, market_value=1.029, reward=6.534e-06, episode=1.022e+04, loss=0.000248\n", + "portfolio_value=1.052, cash_bias=0.2996, market_value=1.052, reward=6.535e-06, episode=1.024e+04, loss=0.000248\n", + "portfolio_value=1.038, cash_bias=0.1923, market_value=1.03, reward=-1.932e-05, episode=1.026e+04, loss=0.000248\n", + "portfolio_value=1.043, cash_bias=0.2892, market_value=1.034, reward=-9.874e-06, episode=1.027e+04, loss=0.0002481\n", + "portfolio_value=1.063, cash_bias=0.2941, market_value=1.042, reward=2.725e-06, episode=1.029e+04, loss=0.000248\n", + "portfolio_value=1.032, cash_bias=0.4402, market_value=1.043, reward=3.558e-05, episode=1.03e+04, loss=0.0002481\n", + "portfolio_value=1.026, cash_bias=0.2291, market_value=1.035, reward=2.186e-05, episode=1.032e+04, loss=0.0002481\n", + "portfolio_value=1.061, cash_bias=0.1834, market_value=1.049, reward=3.11e-05, episode=1.034e+04, loss=0.000248\n", + "portfolio_value=1.053, cash_bias=0.1522, market_value=1.05, reward=3.052e-05, episode=1.035e+04, loss=0.000248\n", + "portfolio_value=1.041, cash_bias=0.2745, market_value=1.028, reward=1.87e-05, episode=1.037e+04, loss=0.000248\n", + "portfolio_value=1.017, cash_bias=0.2927, market_value=1.032, reward=-2.442e-05, episode=1.038e+04, loss=0.0002481\n", + "portfolio_value=1.033, cash_bias=0.2873, market_value=1.019, reward=-1.22e-06, episode=1.04e+04, loss=0.000248\n", + "portfolio_value=1.044, cash_bias=0.219, market_value=1.024, reward=-8.58e-07, episode=1.042e+04, loss=0.000248\n", + "portfolio_value=0.9923, cash_bias=0.2242, market_value=1.031, reward=-7.893e-06, episode=1.043e+04, loss=0.0002481\n", + "portfolio_value=1.044, cash_bias=0.2293, market_value=1.034, reward=1.829e-05, episode=1.045e+04, loss=0.000248\n", + "portfolio_value=1.008, cash_bias=0.2461, market_value=1.01, reward=-2.993e-05, episode=1.046e+04, loss=0.0002481\n", + "portfolio_value=1.051, cash_bias=0.1484, market_value=1.015, reward=1.563e-05, episode=1.048e+04, loss=0.000248\n", + "portfolio_value=1.03, cash_bias=0.2536, market_value=1.009, reward=5.606e-06, episode=1.05e+04, loss=0.000248\n", + "portfolio_value=1.088, cash_bias=0.2449, market_value=1.062, reward=-3.844e-05, episode=1.051e+04, loss=0.000248\n", + "portfolio_value=1.085, cash_bias=0.2738, market_value=1.041, reward=-6.636e-07, episode=1.053e+04, loss=0.000248\n", + "portfolio_value=1.071, cash_bias=0.2307, market_value=1.064, reward=2.09e-05, episode=1.054e+04, loss=0.000248\n", + "portfolio_value=1.053, cash_bias=0.191, market_value=1.043, reward=1.344e-05, episode=1.056e+04, loss=0.000248\n", + "portfolio_value=1.002, cash_bias=0.2206, market_value=1.024, reward=1.717e-05, episode=1.058e+04, loss=0.0002481\n", + "portfolio_value=1.018, cash_bias=0.2058, market_value=0.9974, reward=1.136e-05, episode=1.059e+04, loss=0.0002481\n", + "portfolio_value=1.073, cash_bias=0.258, market_value=1.05, reward=-2.331e-05, episode=1.061e+04, loss=0.000248\n", + "portfolio_value=1.104, cash_bias=0.2397, market_value=1.058, reward=2.444e-05, episode=1.062e+04, loss=0.000248\n", + "portfolio_value=0.9942, cash_bias=0.2293, market_value=1.036, reward=-3.193e-05, episode=1.064e+04, loss=0.0002481\n", + "portfolio_value=1.059, cash_bias=0.2821, market_value=1.061, reward=1.034e-08, episode=1.066e+04, loss=0.000248\n", + "portfolio_value=1.038, cash_bias=0.1189, market_value=1.02, reward=1.864e-05, episode=1.067e+04, loss=0.000248\n", + "portfolio_value=0.9939, cash_bias=0.1471, market_value=1.025, reward=-1.24e-05, episode=1.069e+04, loss=0.0002481\n", + "portfolio_value=1.005, cash_bias=0.3254, market_value=1.017, reward=7.707e-06, episode=1.07e+04, loss=0.0002481\n", + "portfolio_value=1.063, cash_bias=0.1162, market_value=1.052, reward=8.645e-06, episode=1.072e+04, loss=0.000248\n", + "portfolio_value=0.9847, cash_bias=0.3073, market_value=1.021, reward=2.402e-05, episode=1.074e+04, loss=0.0002481\n", + "portfolio_value=1.017, cash_bias=0.286, market_value=1.021, reward=1.118e-05, episode=1.075e+04, loss=0.0002481\n", + "portfolio_value=1.04, cash_bias=0.2114, market_value=1.025, reward=1.207e-05, episode=1.077e+04, loss=0.000248\n", + "portfolio_value=1.014, cash_bias=0.2222, market_value=1.037, reward=-3.907e-05, episode=1.078e+04, loss=0.0002481\n", + "portfolio_value=1.052, cash_bias=0.1455, market_value=1.029, reward=1.053e-05, episode=1.08e+04, loss=0.000248\n", + "portfolio_value=1.011, cash_bias=0.2887, market_value=0.9929, reward=9.711e-06, episode=1.082e+04, loss=0.0002481\n", + "portfolio_value=1.021, cash_bias=0.2106, market_value=1.019, reward=1.038e-05, episode=1.083e+04, loss=0.0002481\n", + "portfolio_value=1.077, cash_bias=0.202, market_value=1.049, reward=-2.847e-06, episode=1.085e+04, loss=0.000248\n", + "portfolio_value=1.057, cash_bias=0.2597, market_value=1.053, reward=1.201e-05, episode=1.086e+04, loss=0.000248\n", + "portfolio_value=1.037, cash_bias=0.3116, market_value=1.039, reward=2.782e-05, episode=1.088e+04, loss=0.000248\n", + "portfolio_value=1.031, cash_bias=0.3667, market_value=1.044, reward=4.213e-06, episode=1.09e+04, loss=0.000248\n", + "portfolio_value=1.04, cash_bias=0.1247, market_value=1.029, reward=1.389e-05, episode=1.091e+04, loss=0.000248\n", + "portfolio_value=1.022, cash_bias=0.31, market_value=1.056, reward=2.045e-05, episode=1.093e+04, loss=0.0002481\n", + "portfolio_value=1.017, cash_bias=0.1644, market_value=1.007, reward=-2.243e-06, episode=1.094e+04, loss=0.0002481\n", + "portfolio_value=1.024, cash_bias=0.4003, market_value=1.036, reward=8.423e-06, episode=1.096e+04, loss=0.0002481\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.053, cash_bias=0.1108, market_value=1.061, reward=2.451e-05, episode=1.098e+04, loss=0.000248\n", + "portfolio_value=1.019, cash_bias=0.3001, market_value=1.033, reward=-1.077e-06, episode=1.099e+04, loss=0.0002481\n", + "portfolio_value=1.038, cash_bias=0.2384, market_value=1.035, reward=6.977e-06, episode=1.101e+04, loss=0.0002484\n", + "portfolio_value=1.038, cash_bias=0.224, market_value=1.027, reward=5.126e-05, episode=1.102e+04, loss=0.000248\n", + "portfolio_value=1.018, cash_bias=0.436, market_value=0.9977, reward=-3.495e-05, episode=1.104e+04, loss=0.000248\n", + "portfolio_value=1.022, cash_bias=0.1419, market_value=1.001, reward=-1.855e-05, episode=1.106e+04, loss=0.000248\n", + "portfolio_value=1.061, cash_bias=0.1943, market_value=1.029, reward=6.192e-06, episode=1.107e+04, loss=0.000248\n", + "portfolio_value=1.105, cash_bias=0.205, market_value=1.076, reward=2.679e-05, episode=1.109e+04, loss=0.000248\n", + "portfolio_value=1.042, cash_bias=0.153, market_value=1.043, reward=1.012e-05, episode=1.11e+04, loss=0.0002482\n", + "portfolio_value=1.097, cash_bias=0.248, market_value=1.055, reward=2.412e-05, episode=1.112e+04, loss=0.0002479\n", + "portfolio_value=1.021, cash_bias=0.1885, market_value=1.033, reward=-2.909e-07, episode=1.114e+04, loss=0.0002481\n", + "portfolio_value=1.042, cash_bias=0.1734, market_value=1.045, reward=-1.649e-05, episode=1.115e+04, loss=0.0002481\n", + "portfolio_value=1.03, cash_bias=0.1238, market_value=1.044, reward=-3.76e-05, episode=1.117e+04, loss=0.000248\n", + "portfolio_value=0.976, cash_bias=0.2773, market_value=1.003, reward=-3.873e-05, episode=1.118e+04, loss=0.0002482\n", + "portfolio_value=1.024, cash_bias=0.2326, market_value=1.031, reward=1.826e-05, episode=1.12e+04, loss=0.0002481\n", + "portfolio_value=1.012, cash_bias=0.1347, market_value=1.031, reward=-3.034e-05, episode=1.122e+04, loss=0.0002481\n", + "portfolio_value=1.012, cash_bias=0.1699, market_value=1.034, reward=1.546e-06, episode=1.123e+04, loss=0.0002481\n", + "portfolio_value=1.051, cash_bias=0.2611, market_value=1.032, reward=2.498e-05, episode=1.125e+04, loss=0.000248\n", + "portfolio_value=1.042, cash_bias=0.1741, market_value=1.059, reward=-2.959e-05, episode=1.126e+04, loss=0.000248\n", + "portfolio_value=1.021, cash_bias=0.3483, market_value=1.036, reward=1.899e-08, episode=1.128e+04, loss=0.0002481\n", + "portfolio_value=1.041, cash_bias=0.4356, market_value=1.027, reward=-2.611e-07, episode=1.13e+04, loss=0.0002481\n", + "portfolio_value=1.026, cash_bias=0.3601, market_value=1.031, reward=-2.635e-05, episode=1.131e+04, loss=0.0002481\n", + "portfolio_value=1.065, cash_bias=0.2088, market_value=1.044, reward=-2.738e-05, episode=1.133e+04, loss=0.000248\n", + "portfolio_value=0.9821, cash_bias=0.2743, market_value=1.03, reward=-2.213e-05, episode=1.134e+04, loss=0.0002481\n", + "portfolio_value=1.062, cash_bias=0.3383, market_value=1.062, reward=1.254e-06, episode=1.136e+04, loss=0.000248\n", + "portfolio_value=1.037, cash_bias=0.2581, market_value=1.028, reward=4.518e-05, episode=1.138e+04, loss=0.0002481\n", + "portfolio_value=1.051, cash_bias=0.325, market_value=1.034, reward=-2.607e-06, episode=1.139e+04, loss=0.000248\n", + "portfolio_value=1.009, cash_bias=0.2562, market_value=1.028, reward=1.353e-05, episode=1.141e+04, loss=0.0002481\n", + "portfolio_value=1.031, cash_bias=0.2091, market_value=1.034, reward=1.382e-05, episode=1.142e+04, loss=0.0002486\n", + "portfolio_value=1.064, cash_bias=0.1842, market_value=1.051, reward=-3.277e-05, episode=1.144e+04, loss=0.000248\n", + "portfolio_value=1.001, cash_bias=0.2547, market_value=1.037, reward=6.94e-06, episode=1.146e+04, loss=0.0002481\n", + "portfolio_value=1.039, cash_bias=0.2218, market_value=1.02, reward=3.19e-05, episode=1.147e+04, loss=0.0002484\n", + "portfolio_value=1.071, cash_bias=0.2296, market_value=1.054, reward=-7.826e-06, episode=1.149e+04, loss=0.000248\n", + "portfolio_value=1.044, cash_bias=0.223, market_value=1.027, reward=9.606e-06, episode=1.15e+04, loss=0.000248\n", + "portfolio_value=1.016, cash_bias=0.1722, market_value=1.021, reward=1.999e-05, episode=1.152e+04, loss=0.0002481\n", + "portfolio_value=1.052, cash_bias=0.1327, market_value=1.057, reward=1.965e-05, episode=1.154e+04, loss=0.0002481\n", + "portfolio_value=1.013, cash_bias=0.2253, market_value=1.018, reward=-2.321e-06, episode=1.155e+04, loss=0.0002481\n", + "portfolio_value=1.045, cash_bias=0.1419, market_value=1.03, reward=7.296e-06, episode=1.157e+04, loss=0.000248\n", + "portfolio_value=1.05, cash_bias=0.1568, market_value=1.032, reward=1.097e-05, episode=1.158e+04, loss=0.0002485\n", + "portfolio_value=1.017, cash_bias=0.1845, market_value=1.053, reward=1.882e-05, episode=1.16e+04, loss=0.0002481\n", + "portfolio_value=1.042, cash_bias=0.2124, market_value=1.04, reward=1.626e-05, episode=1.162e+04, loss=0.000248\n", + "portfolio_value=1.037, cash_bias=0.1854, market_value=1.031, reward=-1.412e-05, episode=1.163e+04, loss=0.0002481\n", + "portfolio_value=1.069, cash_bias=0.2786, market_value=1.051, reward=1.017e-05, episode=1.165e+04, loss=0.000248\n", + "portfolio_value=1.004, cash_bias=0.2361, market_value=1.034, reward=4.032e-06, episode=1.166e+04, loss=0.0002481\n", + "portfolio_value=1.042, cash_bias=0.1535, market_value=1.035, reward=5.971e-06, episode=1.168e+04, loss=0.000248\n", + "portfolio_value=1.031, cash_bias=0.09974, market_value=1.032, reward=-6.584e-06, episode=1.17e+04, loss=0.0002481\n", + "portfolio_value=1.045, cash_bias=0.1224, market_value=1.041, reward=2.766e-05, episode=1.171e+04, loss=0.0002481\n", + "portfolio_value=1.119, cash_bias=0.2202, market_value=1.07, reward=2.123e-05, episode=1.173e+04, loss=0.000248\n", + "portfolio_value=1.053, cash_bias=0.2612, market_value=1.047, reward=-2.1e-05, episode=1.174e+04, loss=0.000248\n", + "portfolio_value=1.036, cash_bias=0.1814, market_value=1.035, reward=-6.801e-06, episode=1.176e+04, loss=0.000248\n", + "portfolio_value=1.013, cash_bias=0.214, market_value=1.011, reward=-1.548e-06, episode=1.178e+04, loss=0.0002481\n", + "portfolio_value=1.053, cash_bias=0.2049, market_value=1.051, reward=1.649e-06, episode=1.179e+04, loss=0.000248\n", + "portfolio_value=1.051, cash_bias=0.1936, market_value=1.049, reward=3.065e-07, episode=1.181e+04, loss=0.000248\n", + "portfolio_value=1.012, cash_bias=0.2239, market_value=1.015, reward=3.964e-07, episode=1.182e+04, loss=0.0002481\n", + "portfolio_value=1.024, cash_bias=0.276, market_value=1.036, reward=3.478e-06, episode=1.184e+04, loss=0.0002484\n", + "portfolio_value=1.037, cash_bias=0.1066, market_value=1.043, reward=3.415e-05, episode=1.186e+04, loss=0.000248\n", + "portfolio_value=1.038, cash_bias=0.3382, market_value=1.053, reward=-6.591e-06, episode=1.187e+04, loss=0.000248\n", + "portfolio_value=1.064, cash_bias=0.3282, market_value=1.045, reward=-7.64e-07, episode=1.189e+04, loss=0.000248\n", + "portfolio_value=1.022, cash_bias=0.2031, market_value=1.026, reward=-1.383e-05, episode=1.19e+04, loss=0.000248\n", + "portfolio_value=1.012, cash_bias=0.1923, market_value=1.043, reward=1.342e-06, episode=1.192e+04, loss=0.0002481\n", + "portfolio_value=1.046, cash_bias=0.1772, market_value=1.049, reward=2.398e-06, episode=1.194e+04, loss=0.000248\n", + "portfolio_value=1.002, cash_bias=0.2513, market_value=1.035, reward=3.055e-06, episode=1.195e+04, loss=0.0002481\n", + "portfolio_value=1.009, cash_bias=0.2506, market_value=1.013, reward=2.649e-05, episode=1.197e+04, loss=0.0002481\n", + "portfolio_value=1.033, cash_bias=0.1375, market_value=1.012, reward=3.072e-06, episode=1.198e+04, loss=0.0002481\n", + "portfolio_value=1.053, cash_bias=0.2487, market_value=1.046, reward=1.76e-05, episode=1.2e+04, loss=0.000248\n", + "portfolio_value=1.027, cash_bias=0.1945, market_value=1.019, reward=-9.087e-06, episode=1.202e+04, loss=0.000248\n", + "portfolio_value=1.035, cash_bias=0.2944, market_value=1.025, reward=-2.268e-05, episode=1.203e+04, loss=0.0002481\n", + "portfolio_value=1.07, cash_bias=0.2609, market_value=1.049, reward=1.223e-07, episode=1.205e+04, loss=0.000248\n", + "portfolio_value=1.08, cash_bias=0.08988, market_value=1.048, reward=-2.168e-05, episode=1.206e+04, loss=0.000248\n", + "portfolio_value=1.024, cash_bias=0.1408, market_value=1.038, reward=-8.255e-06, episode=1.208e+04, loss=0.0002481\n", + "portfolio_value=1.02, cash_bias=0.1734, market_value=1.043, reward=-1.223e-05, episode=1.21e+04, loss=0.0002481\n", + "portfolio_value=1.096, cash_bias=0.2839, market_value=1.058, reward=1.002e-05, episode=1.211e+04, loss=0.000248\n", + "portfolio_value=1.034, cash_bias=0.2508, market_value=1.063, reward=-1.007e-05, episode=1.213e+04, loss=0.000248\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=0.9962, cash_bias=0.2843, market_value=1.01, reward=-5.301e-06, episode=1.214e+04, loss=0.0002481\n", + "portfolio_value=1.094, cash_bias=0.3076, market_value=1.063, reward=2.911e-05, episode=1.216e+04, loss=0.000248\n", + "portfolio_value=1.053, cash_bias=0.2491, market_value=1.04, reward=1.138e-06, episode=1.218e+04, loss=0.000248\n", + "portfolio_value=1.063, cash_bias=0.2369, market_value=1.045, reward=1.51e-05, episode=1.219e+04, loss=0.0002481\n", + "portfolio_value=0.9863, cash_bias=0.1727, market_value=0.9976, reward=-1.552e-07, episode=1.221e+04, loss=0.0002481\n", + "portfolio_value=1.092, cash_bias=0.1192, market_value=1.043, reward=1.641e-05, episode=1.222e+04, loss=0.000248\n", + "portfolio_value=1.056, cash_bias=0.1317, market_value=1.042, reward=-8.997e-06, episode=1.224e+04, loss=0.000248\n", + "portfolio_value=1.03, cash_bias=0.1829, market_value=1.003, reward=5.396e-05, episode=1.226e+04, loss=0.0002481\n", + "portfolio_value=1.077, cash_bias=0.1794, market_value=1.031, reward=-6.247e-06, episode=1.227e+04, loss=0.000248\n", + "portfolio_value=1.039, cash_bias=0.2166, market_value=1.039, reward=-2.138e-06, episode=1.229e+04, loss=0.0002481\n", + "portfolio_value=1.044, cash_bias=0.06596, market_value=1.033, reward=3.621e-05, episode=1.23e+04, loss=0.0002481\n", + "portfolio_value=1.014, cash_bias=0.1692, market_value=1.02, reward=-1.34e-05, episode=1.232e+04, loss=0.0002481\n", + "portfolio_value=1.081, cash_bias=0.1724, market_value=1.047, reward=-1.265e-06, episode=1.234e+04, loss=0.000248\n", + "portfolio_value=1.065, cash_bias=0.2891, market_value=1.03, reward=-1.557e-05, episode=1.235e+04, loss=0.000248\n", + "portfolio_value=0.9992, cash_bias=0.3851, market_value=1.013, reward=1.245e-06, episode=1.237e+04, loss=0.0002481\n", + "portfolio_value=1.063, cash_bias=0.09681, market_value=1.055, reward=-5.356e-06, episode=1.238e+04, loss=0.000248\n", + "portfolio_value=1.087, cash_bias=0.1878, market_value=1.057, reward=1.878e-05, episode=1.24e+04, loss=0.000248\n", + "portfolio_value=1.03, cash_bias=0.2123, market_value=1.049, reward=-2.161e-05, episode=1.242e+04, loss=0.000248\n", + "portfolio_value=1.032, cash_bias=0.1966, market_value=1.022, reward=-1.65e-06, episode=1.243e+04, loss=0.0002481\n", + "portfolio_value=1.005, cash_bias=0.3166, market_value=1.002, reward=-4.883e-06, episode=1.245e+04, loss=0.0002481\n", + "portfolio_value=1.026, cash_bias=0.2283, market_value=1.032, reward=-1.059e-05, episode=1.246e+04, loss=0.000248\n", + "portfolio_value=1.024, cash_bias=0.2019, market_value=0.9973, reward=1.34e-05, episode=1.248e+04, loss=0.0002481\n", + "portfolio_value=1.052, cash_bias=0.3572, market_value=1.039, reward=-1.165e-05, episode=1.25e+04, loss=0.0002481\n", + "portfolio_value=1.019, cash_bias=0.2387, market_value=1.022, reward=1.929e-05, episode=1.251e+04, loss=0.0002481\n", + "portfolio_value=1.045, cash_bias=0.3092, market_value=1.027, reward=3.319e-05, episode=1.253e+04, loss=0.0002481\n", + "portfolio_value=1.018, cash_bias=0.2464, market_value=1.017, reward=2.063e-05, episode=1.254e+04, loss=0.0002481\n", + "portfolio_value=1.087, cash_bias=0.1608, market_value=1.074, reward=3.927e-06, episode=1.256e+04, loss=0.000248\n", + "portfolio_value=1.013, cash_bias=0.312, market_value=1.027, reward=2.195e-05, episode=1.258e+04, loss=0.0002481\n", + "portfolio_value=1.004, cash_bias=0.2192, market_value=1.041, reward=-1.599e-05, episode=1.259e+04, loss=0.0002481\n", + "portfolio_value=1.04, cash_bias=0.153, market_value=1.04, reward=-5.242e-06, episode=1.261e+04, loss=0.0002481\n", + "portfolio_value=1.055, cash_bias=0.1349, market_value=1.032, reward=2.535e-05, episode=1.262e+04, loss=0.0002481\n", + "portfolio_value=1.038, cash_bias=0.2845, market_value=1.055, reward=2.519e-05, episode=1.264e+04, loss=0.0002481\n", + "portfolio_value=1.042, cash_bias=0.2711, market_value=1.03, reward=7.79e-06, episode=1.266e+04, loss=0.000248\n", + "portfolio_value=1.03, cash_bias=0.1374, market_value=1.022, reward=-1.423e-05, episode=1.267e+04, loss=0.000248\n", + "portfolio_value=0.9504, cash_bias=0.1885, market_value=0.9795, reward=-3.796e-05, episode=1.269e+04, loss=0.0002482\n", + "portfolio_value=1.057, cash_bias=0.1591, market_value=1.046, reward=-6.379e-07, episode=1.27e+04, loss=0.000248\n", + "portfolio_value=1.056, cash_bias=0.1772, market_value=1.046, reward=-4.577e-06, episode=1.272e+04, loss=0.0002481\n", + "portfolio_value=1.041, cash_bias=0.1959, market_value=1.032, reward=-2.545e-05, episode=1.274e+04, loss=0.0002481\n", + "portfolio_value=1.03, cash_bias=0.2646, market_value=1.019, reward=1.968e-05, episode=1.275e+04, loss=0.0002481\n", + "portfolio_value=1.022, cash_bias=0.2031, market_value=1.02, reward=1.269e-05, episode=1.277e+04, loss=0.0002481\n", + "portfolio_value=1.017, cash_bias=0.1154, market_value=1.015, reward=1.624e-07, episode=1.278e+04, loss=0.0002481\n", + "portfolio_value=1.077, cash_bias=0.1935, market_value=1.042, reward=2.426e-05, episode=1.28e+04, loss=0.000248\n", + "portfolio_value=1.027, cash_bias=0.2399, market_value=1.025, reward=1.366e-05, episode=1.282e+04, loss=0.000248\n", + "portfolio_value=1.048, cash_bias=0.3391, market_value=1.057, reward=-4.702e-06, episode=1.283e+04, loss=0.000248\n", + "portfolio_value=1.028, cash_bias=0.1176, market_value=1.018, reward=2.348e-06, episode=1.285e+04, loss=0.0002481\n", + "portfolio_value=1.072, cash_bias=0.3996, market_value=1.046, reward=9.881e-06, episode=1.286e+04, loss=0.000248\n", + "portfolio_value=1.023, cash_bias=0.3057, market_value=1.043, reward=6.837e-06, episode=1.288e+04, loss=0.0002481\n", + "portfolio_value=1.005, cash_bias=0.1645, market_value=1.028, reward=1.041e-05, episode=1.29e+04, loss=0.0002481\n", + "portfolio_value=1.004, cash_bias=0.2387, market_value=1.02, reward=1.945e-05, episode=1.291e+04, loss=0.0002481\n", + "portfolio_value=1.092, cash_bias=0.2588, market_value=1.065, reward=-6.722e-06, episode=1.293e+04, loss=0.000248\n", + "portfolio_value=1.035, cash_bias=0.2689, market_value=1.034, reward=2.195e-05, episode=1.294e+04, loss=0.0002481\n", + "portfolio_value=1.045, cash_bias=0.2757, market_value=1.031, reward=1.113e-05, episode=1.296e+04, loss=0.000248\n", + "portfolio_value=0.9992, cash_bias=0.2803, market_value=1.011, reward=1.513e-06, episode=1.298e+04, loss=0.0002481\n", + "portfolio_value=1.026, cash_bias=0.2735, market_value=1.029, reward=-2.237e-05, episode=1.299e+04, loss=0.000248\n", + "portfolio_value=1.023, cash_bias=0.3124, market_value=1.031, reward=-1.381e-05, episode=1.301e+04, loss=0.0002481\n", + "portfolio_value=1.026, cash_bias=0.1391, market_value=1.026, reward=1.351e-05, episode=1.302e+04, loss=0.0002481\n", + "portfolio_value=1.027, cash_bias=0.1377, market_value=1.045, reward=5.622e-06, episode=1.304e+04, loss=0.000248\n", + "portfolio_value=1.025, cash_bias=0.142, market_value=1.05, reward=-1.557e-05, episode=1.306e+04, loss=0.0002481\n", + "portfolio_value=1.035, cash_bias=0.09189, market_value=1.038, reward=-1.459e-05, episode=1.307e+04, loss=0.000248\n", + "portfolio_value=0.9699, cash_bias=0.2426, market_value=0.9766, reward=-1.004e-05, episode=1.309e+04, loss=0.0002487\n", + "portfolio_value=1.018, cash_bias=0.3478, market_value=1.039, reward=-1.457e-06, episode=1.31e+04, loss=0.0002481\n", + "portfolio_value=1.043, cash_bias=0.3007, market_value=1.027, reward=1.311e-05, episode=1.312e+04, loss=0.000248\n", + "portfolio_value=1.043, cash_bias=0.2029, market_value=1.037, reward=9.072e-06, episode=1.314e+04, loss=0.000248\n", + "portfolio_value=1.061, cash_bias=0.1504, market_value=1.033, reward=1.074e-05, episode=1.315e+04, loss=0.000248\n", + "portfolio_value=1.016, cash_bias=0.1411, market_value=1.022, reward=-1.275e-05, episode=1.317e+04, loss=0.0002481\n", + "portfolio_value=1.035, cash_bias=0.1362, market_value=1.039, reward=-4.947e-07, episode=1.318e+04, loss=0.0002481\n", + "portfolio_value=1.086, cash_bias=0.1826, market_value=1.068, reward=-9.966e-06, episode=1.32e+04, loss=0.000248\n", + "portfolio_value=1.084, cash_bias=0.1751, market_value=1.071, reward=-5.348e-05, episode=1.322e+04, loss=0.000248\n", + "portfolio_value=1.014, cash_bias=0.1611, market_value=1.007, reward=7.036e-06, episode=1.323e+04, loss=0.0002483\n", + "portfolio_value=1.031, cash_bias=0.2492, market_value=1.026, reward=1.05e-05, episode=1.325e+04, loss=0.0002481\n", + "portfolio_value=1.047, cash_bias=0.229, market_value=1.027, reward=1.964e-05, episode=1.326e+04, loss=0.0002481\n", + "portfolio_value=1.031, cash_bias=0.1984, market_value=1.064, reward=-1.585e-05, episode=1.328e+04, loss=0.0002481\n", + "portfolio_value=1.008, cash_bias=0.2826, market_value=1.03, reward=6.906e-06, episode=1.33e+04, loss=0.0002481\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.016, cash_bias=0.2811, market_value=1.027, reward=-2.855e-06, episode=1.331e+04, loss=0.000248\n", + "portfolio_value=1.019, cash_bias=0.1379, market_value=1.046, reward=-2.49e-05, episode=1.333e+04, loss=0.0002481\n", + "portfolio_value=1.051, cash_bias=0.2097, market_value=1.031, reward=-1.056e-05, episode=1.334e+04, loss=0.000248\n", + "portfolio_value=1.022, cash_bias=0.1302, market_value=1.002, reward=5.876e-06, episode=1.336e+04, loss=0.0002481\n", + "portfolio_value=1.013, cash_bias=0.2979, market_value=1.017, reward=-3.207e-06, episode=1.338e+04, loss=0.0002481\n", + "portfolio_value=1.052, cash_bias=0.3713, market_value=1.045, reward=2.158e-05, episode=1.339e+04, loss=0.000248\n", + "portfolio_value=1.01, cash_bias=0.2456, market_value=1.005, reward=4.504e-06, episode=1.341e+04, loss=0.0002482\n", + "portfolio_value=1.011, cash_bias=0.4574, market_value=1.009, reward=1.666e-05, episode=1.342e+04, loss=0.0002481\n", + "portfolio_value=1.035, cash_bias=0.2843, market_value=1.015, reward=9.034e-06, episode=1.344e+04, loss=0.000248\n", + "portfolio_value=1.014, cash_bias=0.2661, market_value=1.012, reward=-6.108e-07, episode=1.346e+04, loss=0.0002481\n", + "portfolio_value=1.032, cash_bias=0.2658, market_value=1.043, reward=7.372e-06, episode=1.347e+04, loss=0.0002481\n", + "portfolio_value=0.9961, cash_bias=0.1262, market_value=1.031, reward=3.435e-06, episode=1.349e+04, loss=0.0002481\n", + "portfolio_value=1.062, cash_bias=0.332, market_value=1.028, reward=1.716e-05, episode=1.35e+04, loss=0.000248\n", + "portfolio_value=1.035, cash_bias=0.3446, market_value=1.043, reward=2.749e-05, episode=1.352e+04, loss=0.0002481\n", + "portfolio_value=1.022, cash_bias=0.3204, market_value=1.032, reward=-1.743e-06, episode=1.354e+04, loss=0.0002481\n", + "portfolio_value=1.066, cash_bias=0.2054, market_value=1.066, reward=6.003e-06, episode=1.355e+04, loss=0.000248\n", + "portfolio_value=1.025, cash_bias=0.1653, market_value=1.021, reward=-1.372e-06, episode=1.357e+04, loss=0.0002481\n", + "portfolio_value=1.01, cash_bias=0.218, market_value=1.018, reward=2.862e-05, episode=1.358e+04, loss=0.0002481\n", + "portfolio_value=1.068, cash_bias=0.1282, market_value=1.034, reward=-2.266e-05, episode=1.36e+04, loss=0.000248\n", + "portfolio_value=1.037, cash_bias=0.1836, market_value=1.031, reward=7.064e-06, episode=1.362e+04, loss=0.000248\n", + "portfolio_value=1.054, cash_bias=0.2983, market_value=1.052, reward=1.04e-05, episode=1.363e+04, loss=0.0002481\n", + "portfolio_value=0.9977, cash_bias=0.3, market_value=1.021, reward=-1.488e-05, episode=1.365e+04, loss=0.0002481\n", + "portfolio_value=1.047, cash_bias=0.1891, market_value=1.034, reward=-1.941e-05, episode=1.366e+04, loss=0.000248\n", + "portfolio_value=1.046, cash_bias=0.2523, market_value=1.049, reward=-1.998e-06, episode=1.368e+04, loss=0.0002481\n", + "portfolio_value=0.9864, cash_bias=0.1741, market_value=1.006, reward=8.368e-06, episode=1.37e+04, loss=0.0002481\n", + "portfolio_value=1.048, cash_bias=0.1389, market_value=1.041, reward=-2.668e-06, episode=1.371e+04, loss=0.0002481\n", + "portfolio_value=1.054, cash_bias=0.1578, market_value=1.045, reward=-3.014e-05, episode=1.373e+04, loss=0.000248\n", + "portfolio_value=1.053, cash_bias=0.2438, market_value=1.035, reward=-1.077e-05, episode=1.374e+04, loss=0.000248\n", + "portfolio_value=1.061, cash_bias=0.1936, market_value=1.019, reward=-2.666e-05, episode=1.376e+04, loss=0.000248\n", + "portfolio_value=0.9683, cash_bias=0.2812, market_value=0.9889, reward=-6.315e-06, episode=1.378e+04, loss=0.0002481\n", + "portfolio_value=0.9855, cash_bias=0.2608, market_value=1.017, reward=-8.97e-06, episode=1.379e+04, loss=0.0002481\n", + "portfolio_value=0.978, cash_bias=0.2239, market_value=1.01, reward=-5.004e-06, episode=1.381e+04, loss=0.0002481\n", + "portfolio_value=1.051, cash_bias=0.3465, market_value=1.061, reward=1.741e-05, episode=1.382e+04, loss=0.000248\n", + "portfolio_value=1.055, cash_bias=0.1448, market_value=1.045, reward=3.181e-05, episode=1.384e+04, loss=0.000248\n", + "portfolio_value=1.064, cash_bias=0.2758, market_value=1.048, reward=-1.689e-05, episode=1.386e+04, loss=0.000248\n", + "portfolio_value=1.07, cash_bias=0.1343, market_value=1.066, reward=1.444e-05, episode=1.387e+04, loss=0.0002479\n", + "portfolio_value=1.034, cash_bias=0.3648, market_value=1.023, reward=-1.091e-05, episode=1.389e+04, loss=0.000248\n", + "portfolio_value=1.017, cash_bias=0.1642, market_value=1.024, reward=1.756e-05, episode=1.39e+04, loss=0.0002481\n", + "portfolio_value=0.9934, cash_bias=0.2133, market_value=0.9953, reward=3.089e-06, episode=1.392e+04, loss=0.0002481\n", + "portfolio_value=1.064, cash_bias=0.1858, market_value=1.04, reward=-1.082e-05, episode=1.394e+04, loss=0.000248\n", + "portfolio_value=1.035, cash_bias=0.2441, market_value=1.016, reward=1.878e-06, episode=1.395e+04, loss=0.0002482\n", + "portfolio_value=1.06, cash_bias=0.3873, market_value=1.04, reward=1.583e-05, episode=1.397e+04, loss=0.000248\n", + "portfolio_value=1.017, cash_bias=0.3138, market_value=1.008, reward=-1.197e-05, episode=1.398e+04, loss=0.0002481\n", + "portfolio_value=1.074, cash_bias=0.1979, market_value=1.042, reward=2.155e-05, episode=1.4e+04, loss=0.000248\n", + "portfolio_value=1.017, cash_bias=0.3074, market_value=1.025, reward=1.898e-05, episode=1.402e+04, loss=0.0002481\n", + "portfolio_value=1.059, cash_bias=0.2623, market_value=1.049, reward=2.179e-06, episode=1.403e+04, loss=0.000248\n", + "portfolio_value=1.056, cash_bias=0.1069, market_value=1.061, reward=-1.603e-05, episode=1.405e+04, loss=0.000248\n", + "portfolio_value=1.073, cash_bias=0.1443, market_value=1.063, reward=-1.342e-05, episode=1.406e+04, loss=0.000248\n", + "portfolio_value=1.044, cash_bias=0.1331, market_value=1.032, reward=1.346e-05, episode=1.408e+04, loss=0.0002481\n", + "portfolio_value=1.032, cash_bias=0.1858, market_value=1.042, reward=1.172e-05, episode=1.41e+04, loss=0.0002481\n", + "portfolio_value=0.9789, cash_bias=0.2021, market_value=1.012, reward=5.032e-05, episode=1.411e+04, loss=0.0002482\n", + "portfolio_value=1.014, cash_bias=0.2039, market_value=0.9971, reward=-1.237e-05, episode=1.413e+04, loss=0.000248\n", + "portfolio_value=0.9726, cash_bias=0.3216, market_value=1.007, reward=-2.238e-05, episode=1.414e+04, loss=0.0002481\n", + "portfolio_value=1.081, cash_bias=0.2232, market_value=1.053, reward=-1.538e-05, episode=1.416e+04, loss=0.000248\n", + "portfolio_value=1.083, cash_bias=0.2619, market_value=1.048, reward=-4.016e-06, episode=1.418e+04, loss=0.000248\n", + "portfolio_value=1.024, cash_bias=0.2563, market_value=1.015, reward=8.939e-06, episode=1.419e+04, loss=0.000248\n", + "portfolio_value=1.07, cash_bias=0.269, market_value=1.053, reward=6.966e-06, episode=1.421e+04, loss=0.000248\n", + "portfolio_value=1.043, cash_bias=0.2483, market_value=1.043, reward=5.506e-06, episode=1.422e+04, loss=0.000248\n", + "portfolio_value=1.011, cash_bias=0.2002, market_value=1.032, reward=9.548e-06, episode=1.424e+04, loss=0.0002481\n", + "portfolio_value=1.044, cash_bias=0.1618, market_value=1.018, reward=1.28e-06, episode=1.426e+04, loss=0.0002481\n", + "portfolio_value=1.054, cash_bias=0.1995, market_value=1.029, reward=2.112e-06, episode=1.427e+04, loss=0.000248\n", + "portfolio_value=1.017, cash_bias=0.28, market_value=1.016, reward=-2.838e-05, episode=1.429e+04, loss=0.000248\n", + "portfolio_value=1.03, cash_bias=0.3257, market_value=1.026, reward=-1.196e-05, episode=1.43e+04, loss=0.000248\n", + "portfolio_value=1.076, cash_bias=0.2111, market_value=1.057, reward=1.763e-05, episode=1.432e+04, loss=0.000248\n", + "portfolio_value=1.094, cash_bias=0.1783, market_value=1.036, reward=2.038e-05, episode=1.434e+04, loss=0.000248\n", + "portfolio_value=0.9961, cash_bias=0.14, market_value=1.013, reward=5.576e-06, episode=1.435e+04, loss=0.000248\n", + "portfolio_value=1.041, cash_bias=0.1942, market_value=1.03, reward=1.578e-05, episode=1.437e+04, loss=0.000248\n", + "portfolio_value=1.055, cash_bias=0.2328, market_value=1.043, reward=5.99e-06, episode=1.438e+04, loss=0.000248\n", + "portfolio_value=1.065, cash_bias=0.1076, market_value=1.067, reward=-4.918e-05, episode=1.44e+04, loss=0.000248\n", + "portfolio_value=1.033, cash_bias=0.09726, market_value=1.026, reward=-2.773e-05, episode=1.442e+04, loss=0.000248\n", + "portfolio_value=1.05, cash_bias=0.08696, market_value=1.04, reward=1.437e-05, episode=1.443e+04, loss=0.000248\n", + "portfolio_value=1.004, cash_bias=0.1208, market_value=1.014, reward=3.308e-06, episode=1.445e+04, loss=0.0002481\n", + "portfolio_value=1.078, cash_bias=0.2076, market_value=1.062, reward=6.543e-06, episode=1.446e+04, loss=0.000248\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "portfolio_value=1.014, cash_bias=0.3171, market_value=1.009, reward=6.229e-06, episode=1.448e+04, loss=0.0002481\n", + "portfolio_value=1.023, cash_bias=0.1674, market_value=1.031, reward=9.427e-06, episode=1.45e+04, loss=0.0002481\n", + "portfolio_value=0.9927, cash_bias=0.2508, market_value=1.007, reward=-8.455e-08, episode=1.451e+04, loss=0.0002481\n", + "portfolio_value=1.034, cash_bias=0.208, market_value=1.022, reward=-2.309e-05, episode=1.453e+04, loss=0.000248\n", + "portfolio_value=1.031, cash_bias=0.167, market_value=1.045, reward=4.485e-06, episode=1.454e+04, loss=0.000248\n", + "portfolio_value=1.07, cash_bias=0.2217, market_value=1.039, reward=-1.757e-05, episode=1.456e+04, loss=0.000248\n", + "portfolio_value=1.037, cash_bias=0.2026, market_value=1.043, reward=1.725e-05, episode=1.458e+04, loss=0.000248\n", + "portfolio_value=1.042, cash_bias=0.1602, market_value=1.031, reward=1.916e-05, episode=1.459e+04, loss=0.0002481\n", + "portfolio_value=1.049, cash_bias=0.3001, market_value=1.016, reward=-2.504e-05, episode=1.461e+04, loss=0.000248\n", + "portfolio_value=0.999, cash_bias=0.1716, market_value=0.9896, reward=2.461e-07, episode=1.462e+04, loss=0.0002481\n", + "portfolio_value=1.041, cash_bias=0.119, market_value=1.027, reward=-2.742e-05, episode=1.464e+04, loss=0.000248\n", + "portfolio_value=1.001, cash_bias=0.2774, market_value=1.019, reward=-3.716e-05, episode=1.466e+04, loss=0.0002481\n", + "portfolio_value=1.033, cash_bias=0.07415, market_value=1.038, reward=-2.689e-05, episode=1.467e+04, loss=0.000248\n", + "portfolio_value=0.9972, cash_bias=0.1933, market_value=0.9975, reward=5.87e-07, episode=1.469e+04, loss=0.0002484\n", + "portfolio_value=1.044, cash_bias=0.3146, market_value=1.031, reward=9.197e-07, episode=1.47e+04, loss=0.0002481\n", + "portfolio_value=1.041, cash_bias=0.2969, market_value=1.045, reward=1.599e-05, episode=1.472e+04, loss=0.000248\n", + "portfolio_value=1.022, cash_bias=0.261, market_value=1.041, reward=1.598e-05, episode=1.474e+04, loss=0.0002481\n", + "portfolio_value=1.049, cash_bias=0.3136, market_value=1.044, reward=-3.73e-06, episode=1.475e+04, loss=0.000248\n", + "portfolio_value=1.001, cash_bias=0.1898, market_value=1.01, reward=6.705e-06, episode=1.477e+04, loss=0.0002481\n", + "portfolio_value=1.076, cash_bias=0.07276, market_value=1.044, reward=-2.283e-05, episode=1.478e+04, loss=0.000248\n", + "portfolio_value=1.047, cash_bias=0.1367, market_value=1.052, reward=-1.168e-06, episode=1.48e+04, loss=0.000248\n", + "portfolio_value=1.109, cash_bias=0.1434, market_value=1.088, reward=2.438e-05, episode=1.482e+04, loss=0.000248\n", + "portfolio_value=0.9868, cash_bias=0.2442, market_value=1.006, reward=-3.935e-07, episode=1.483e+04, loss=0.0002481\n", + "portfolio_value=1.047, cash_bias=0.2489, market_value=1.024, reward=2.042e-06, episode=1.485e+04, loss=0.000248\n", + "portfolio_value=1.058, cash_bias=0.2302, market_value=1.041, reward=-8.791e-07, episode=1.486e+04, loss=0.000248\n", + "portfolio_value=1.012, cash_bias=0.307, market_value=1.035, reward=-7.448e-06, episode=1.488e+04, loss=0.0002481\n", + "portfolio_value=1.071, cash_bias=0.2585, market_value=1.065, reward=6.61e-06, episode=1.49e+04, loss=0.000248\n", + "portfolio_value=0.9926, cash_bias=0.2128, market_value=1.049, reward=-4.03e-05, episode=1.491e+04, loss=0.0002481\n", + "portfolio_value=1.025, cash_bias=0.2025, market_value=1.021, reward=1.87e-05, episode=1.493e+04, loss=0.0002481\n", + "portfolio_value=1.07, cash_bias=0.1812, market_value=1.042, reward=1.303e-05, episode=1.494e+04, loss=0.0002481\n", + "portfolio_value=1.019, cash_bias=0.1399, market_value=1.016, reward=3.216e-06, episode=1.496e+04, loss=0.0002481\n", + "portfolio_value=1.029, cash_bias=0.2203, market_value=1.002, reward=5.403e-05, episode=1.498e+04, loss=0.0002484\n", + "portfolio_value=0.9767, cash_bias=0.2149, market_value=0.9967, reward=-2.998e-06, episode=1.499e+04, loss=0.0002481\n", + "portfolio_value=0.965, cash_bias=0.1541, market_value=1.018, reward=-7.172e-06, episode=1.501e+04, loss=0.0002484\n", + "portfolio_value=1.045, cash_bias=0.1433, market_value=1.04, reward=-3.95e-06, episode=1.502e+04, loss=0.000248\n", + "portfolio_value=1.112, cash_bias=0.1873, market_value=1.07, reward=1.424e-05, episode=1.504e+04, loss=0.000248\n", + "portfolio_value=1.012, cash_bias=0.1693, market_value=1.014, reward=2.055e-05, episode=1.506e+04, loss=0.0002481\n", + "portfolio_value=1.008, cash_bias=0.1524, market_value=1.011, reward=-4.818e-06, episode=1.507e+04, loss=0.0002481\n", + "portfolio_value=1.043, cash_bias=0.2202, market_value=1.045, reward=7.389e-06, episode=1.509e+04, loss=0.0002481\n", + "portfolio_value=1.037, cash_bias=0.1301, market_value=1.039, reward=-9.791e-06, episode=1.51e+04, loss=0.0002481\n", + "portfolio_value=1.017, cash_bias=0.2978, market_value=1.004, reward=2.327e-05, episode=1.512e+04, loss=0.0002481\n", + "portfolio_value=1.014, cash_bias=0.2635, market_value=1.061, reward=-2.799e-05, episode=1.514e+04, loss=0.000248\n", + "portfolio_value=1.044, cash_bias=0.1994, market_value=1.025, reward=1.795e-07, episode=1.515e+04, loss=0.000248\n", + "portfolio_value=1.045, cash_bias=0.1366, market_value=1.02, reward=3.765e-05, episode=1.517e+04, loss=0.000248\n", + "portfolio_value=1.002, cash_bias=0.2602, market_value=1.017, reward=1.129e-05, episode=1.518e+04, loss=0.0002481\n", + "portfolio_value=1.049, cash_bias=0.2035, market_value=1.029, reward=1.022e-06, episode=1.52e+04, loss=0.0002486\n", + "portfolio_value=1.025, cash_bias=0.3086, market_value=1.035, reward=-2.742e-05, episode=1.522e+04, loss=0.0002481\n", + "portfolio_value=1.059, cash_bias=0.2571, market_value=1.03, reward=-1.804e-05, episode=1.523e+04, loss=0.000248\n", + "portfolio_value=1.035, cash_bias=0.1706, market_value=1.037, reward=4.758e-06, episode=1.525e+04, loss=0.0002481\n", + "portfolio_value=1.057, cash_bias=0.1317, market_value=1.04, reward=6.179e-06, episode=1.526e+04, loss=0.000248\n", + "portfolio_value=0.9988, cash_bias=0.267, market_value=1.023, reward=-1.754e-05, episode=1.528e+04, loss=0.0002481\n", + "portfolio_value=1.062, cash_bias=0.1944, market_value=1.03, reward=4.756e-05, episode=1.53e+04, loss=0.000248\n", + "portfolio_value=1.035, cash_bias=0.07456, market_value=1.021, reward=9.208e-06, episode=1.531e+04, loss=0.000248\n", + "portfolio_value=1.07, cash_bias=0.3164, market_value=1.041, reward=5.517e-06, episode=1.533e+04, loss=0.0002481\n", + "portfolio_value=1.056, cash_bias=0.2407, market_value=1.044, reward=6.8e-06, episode=1.534e+04, loss=0.000248\n", + "portfolio_value=0.9515, cash_bias=0.2795, market_value=0.9838, reward=-1.899e-05, episode=1.536e+04, loss=0.0002482\n", + "portfolio_value=1.067, cash_bias=0.2276, market_value=1.037, reward=3.458e-06, episode=1.538e+04, loss=0.000248\n", + "portfolio_value=1.046, cash_bias=0.2038, market_value=1.029, reward=-1.17e-05, episode=1.539e+04, loss=0.0002481\n", + "portfolio_value=1.033, cash_bias=0.2268, market_value=1.031, reward=2.578e-06, episode=1.541e+04, loss=0.000248\n", + "portfolio_value=1.057, cash_bias=0.208, market_value=1.048, reward=-4.996e-06, episode=1.542e+04, loss=0.000248\n", + "portfolio_value=0.9837, cash_bias=0.2493, market_value=1.037, reward=-2.948e-06, episode=1.544e+04, loss=0.0002481\n", + "portfolio_value=1.053, cash_bias=0.1684, market_value=1.071, reward=1.561e-05, episode=1.546e+04, loss=0.000248\n", + "portfolio_value=1.03, cash_bias=0.2158, market_value=1.023, reward=-2.197e-05, episode=1.547e+04, loss=0.000248\n", + "portfolio_value=1.015, cash_bias=0.2614, market_value=1.019, reward=3.821e-05, episode=1.549e+04, loss=0.0002485\n", + "portfolio_value=1.012, cash_bias=0.2604, market_value=1.038, reward=9.712e-06, episode=1.55e+04, loss=0.0002487\n", + "portfolio_value=1.013, cash_bias=0.2212, market_value=1.028, reward=2.284e-05, episode=1.552e+04, loss=0.0002489\n", + "portfolio_value=1.027, cash_bias=0.2953, market_value=1.021, reward=6.177e-06, episode=1.554e+04, loss=0.0002491\n", + "portfolio_value=1.015, cash_bias=0.1895, market_value=1.021, reward=-1.789e-05, episode=1.555e+04, loss=0.0002491\n", + "portfolio_value=1.055, cash_bias=0.3367, market_value=1.055, reward=1.713e-05, episode=1.557e+04, loss=0.000249\n", + "portfolio_value=1.056, cash_bias=0.2078, market_value=1.019, reward=3.375e-05, episode=1.558e+04, loss=0.0002491\n", + "portfolio_value=1.051, cash_bias=0.2311, market_value=1.029, reward=7.167e-06, episode=1.56e+04, loss=0.0002491\n", + "portfolio_value=1.005, cash_bias=0.1088, market_value=1.028, reward=7.279e-06, episode=1.562e+04, loss=0.0002492\n", + "portfolio_value=0.9757, cash_bias=0.1734, market_value=1.018, reward=-2.137e-07, episode=1.563e+04, loss=0.0002492\n", + "\n" + ] + } + ], + "source": [ + "memory = ReplayMemory(params.num_steps)\n", + "# memory = PrioritisedReplayMemory(params.num_steps)\n", + "\n", + "num_inputs = int(np.prod(env.observation_space.shape))\n", + "num_outputs = int(np.prod(env.action_space.shape))\n", + "\n", + "state = env.reset()\n", + "state = Variable(torch.Tensor(state).unsqueeze(0))\n", + "done = True\n", + "episode_length = 0\n", + "reports = []\n", + "\n", + "with tqdm(total=params.time_horizon, mininterval=2, unit='steps') as p:\n", + " episode = -1 \n", + " steps = 0\n", + " # horizon loop\n", + " while steps < params.time_horizon:\n", + " infos = []\n", + " episode_length = 0\n", + " # Sample data from the policy\n", + " while (len(memory.memory) < params.num_steps):\n", + " states = []\n", + " actions = []\n", + " rewards = []\n", + " values = []\n", + " returns = []\n", + " advantages = []\n", + " av_reward = 0\n", + " cum_reward = 0\n", + " cum_done = 0\n", + " # n steps loops\n", + " for step in range(params.num_steps):\n", + " # shared_obs_stats.observes(state)\n", + " # state = shared_obs_stats.normalize(state)\n", + " states.append(state)\n", + " \n", + " mu, sigma_sq, v = model(state)\n", + " eps = torch.randn(mu.size())\n", + " action = (mu + sigma_sq.sqrt() * Variable(eps))\n", + " env_action = action.data.squeeze().numpy()\n", + " state, reward, done, info = env.step(env_action)\n", + " done = (done or episode_length >= params.max_episode_length)\n", + " \n", + " cum_reward += reward\n", + " reward = max(min(reward, 1), -1)\n", + " rewards.append(reward)\n", + " actions.append(action)\n", + " values.append(v)\n", + " \n", + " steps+=1 \n", + " p.update(1)\n", + " if done:\n", + " episode += 1\n", + " cum_done += 1\n", + " av_reward += cum_reward\n", + " p.desc='av_reward={: 2.8f}'.format(av_reward / float(cum_done))\n", + " cum_reward = 0\n", + " episode_length = 0\n", + " infos.append(info)\n", + " state = env.reset()\n", + " \n", + " state = Variable(torch.Tensor(state).unsqueeze(0))\n", + " \n", + " if done:\n", + " break\n", + " \n", + " # one last step\n", + " R = torch.zeros(1, 1)\n", + " if not done:\n", + " _, _, v = model(state)\n", + " R = v.data\n", + " \n", + " # compute returns and GAE(lambda) advantages:\n", + " values.append(Variable(R))\n", + " R = Variable(R)\n", + " A = Variable(torch.zeros(1, 1))\n", + " for i in reversed(range(len(rewards))):\n", + " td = rewards[i] + params.gamma*values[i+1].data[0,0] - values[i].data[0,0]\n", + " A = float(td) + params.gamma * params.gae_param * A\n", + " advantages.insert(0, A)\n", + " R = A + values[i]\n", + " returns.insert(0, R)\n", + " \n", + " # store useful info:\n", + " memory.push([states, actions, returns, advantages])\n", + " \n", + "\n", + " # perform several epochs of optimization on the sampled data\n", + " model_old = Model(env.observation_space.shape,\n", + " env.action_space.shape)\n", + " model_old.load_state_dict(model.state_dict())\n", + " av_loss = 0\n", + " for k in range(params.num_epoch):\n", + " # cf https://github.com/openai/baselines/blob/master/baselines/pposgd/pposgd_simple.py\n", + " batch_states, batch_actions, batch_returns, batch_advantages = memory.sample(\n", + " params.batch_size)\n", + " \n", + " # old probas\n", + " mu_old, sigma_sq_old, v_pred_old = model_old(batch_states.detach())\n", + " probs_old = normal(batch_actions, mu_old, sigma_sq_old)\n", + " \n", + " # new probas\n", + " mu, sigma_sq, v_pred = model(batch_states)\n", + " probs = normal(batch_actions, mu, sigma_sq)\n", + " \n", + " # ratio\n", + " ratio = probs / (1e-15 + probs_old)\n", + " \n", + " # surrogate clip loss\n", + " surr1 = ratio * torch.cat([batch_advantages]*num_outputs,1) # surrogate from conservative policy iteration\n", + " surr2 = ratio.clamp(1-params.clip, 1+params.clip) * torch.cat([batch_advantages]*num_outputs,1)\n", + " loss_clip = -torch.mean(torch.min(surr1, surr2))\n", + " # should this be a mean along axis 0?\n", + " \n", + " # state-value function loss, do we even need this if they don't share params?\n", + " vfloss1 = (v_pred - batch_returns)**2\n", + " v_pred_clipped = v_pred_old + (v_pred - v_pred_old).clamp(-params.clip, params.clip)\n", + " vfloss2 = (v_pred_clipped - batch_returns)**2\n", + " loss_value = 0.5 * torch.mean(torch.max(vfloss1, vfloss2))\n", + " # should this be a mean along axis 0?\n", + " \n", + " # loss on entropy bonus to ensure sufficient exploration\n", + " loss_ent = -params.ent_coeff*torch.mean(probs*torch.log(probs+1e-5))\n", + " \n", + " # total\n", + " total_loss = (loss_clip + loss_value + loss_ent)\n", + "# total_loss = (loss_clip - loss_value + loss_ent)\n", + " av_loss += loss_value.data[0] / float(params.num_epoch)\n", + " \n", + " # before step, update old_model:\n", + " model_old.load_state_dict(model.state_dict())\n", + " \n", + " # step\n", + " optimizer.zero_grad()\n", + " total_loss.backward(retain_variables=True)\n", + " optimizer.step()\n", + " \n", + " # t finish, print:\n", + " df_infos = pd.DataFrame(infos)\n", + " \n", + " # show stats?\n", + "# display(df_infos[[\"cash_bias\",\"return\",\"portfolio_value\",\"market_value\"]].describe().loc[[\"min\",\"mean\",\"max\"]])\n", + " \n", + " report=OrderedDict(\n", + " episode=episode,\n", + "# reward=av_reward / float(cum_done),\n", + " loss=av_loss,\n", + " cash_bias=df_infos.cash_bias.mean(),\n", + " market_value=df_infos.market_value.mean(),\n", + " portfolio_value=df_infos.portfolio_value.mean(),\n", + " reward=df_infos.reward.mean()\n", + " )\n", + " \n", + " s = ', '.join(['{}={:2.4g}'.format(key,value) for key,value in report.items()])\n", + " print(s)\n", + " \n", + " reports.append(report)\n", + " \n", + " memory.clear()\n", + " torch.save(model_old, save_path)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T20:16:30.623423Z", + "start_time": "2017-08-05T20:16:29.869596Z" + } + }, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAuEAAALICAYAAADR4y5sAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# show progress\n", + "df=pd.DataFrame(reports)\n", + "g = sns.jointplot(x=\"episode\", y=\"loss\", data=df, kind=\"reg\", size=10)\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T22:53:19.502553Z", + "start_time": "2017-08-05T22:53:08.630362Z" + }, + "scrolled": false + }, + "outputs": [ + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEyCAYAAADgEkc1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEyCAYAAAD9QLvrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCCG8jCR+IYTwMv8ftosd9kWU9bQAAAAASUVORK5CYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEyCAYAAADzxQfBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQrgpCQBCCOGm/j9A2HZ0y0bxpAAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEyCAYAAAAY8rzCAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEyCAYAAAAWdwDoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAtpEFtPSUE19AAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEyCAYAAAACzlxMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIYRoMAkeQgghGkyChxBCiAb7/7uAaBFdACOQAAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAEyCAYAAAAfnKCSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/wedOTpKVUyA7wAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEyCAYAAAD9QLvrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nfolEIvExpPBLJBKJjyGFXyKRSHyM/wfmJsosxx65LAAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEyCAYAAADzxQfBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n9AAAAABJRU5ErkJggg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEyCAYAAAAWdwDoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# env_test = env\n", + "\n", + "# Test\n", + "for i in range(10):\n", + " model.train(False)\n", + " state = env_test.reset()\n", + " for i in range(250):\n", + " state = Variable(torch.Tensor(state).unsqueeze(0))\n", + " mu, sigma_sq, v = model(state)\n", + " eps = torch.randn(mu.size())\n", + " action = (mu + sigma_sq.sqrt() * Variable(eps))\n", + " env_action = action.data.squeeze().numpy()\n", + " state, reward, done, info = env_test.step(env_action)\n", + " if done:\n", + " break\n", + "\n", + " env_test.plot()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T08:03:17.463005Z", + "start_time": "2017-08-05T08:02:49.857Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T06:59:47.486363Z", + "start_time": "2017-08-05T06:59:47.404265Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "ExecuteTime": { + "end_time": "2017-08-05T06:55:09.249081Z", + "start_time": "2017-08-05T06:55:09.199392Z" + }, + "collapsed": true + }, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "jupyter3", + "language": "python", + "name": "jupyter3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.5.3" + }, + "toc": { + "colors": { + "hover_highlight": "#DAA520", + "navigate_num": "#000000", + "navigate_text": "#333333", + "running_highlight": "#FF0000", + "selected_highlight": "#FFD700", + "sidebar_border": "#EEEEEE", + "wrapper_background": "#FFFFFF" + }, + "moveMenuLeft": true, + "nav_menu": { + "height": "85px", + "width": "252px" + }, + "navigate_menu": true, + "number_sections": true, + "sideBar": true, + "threshold": 4, + "toc_cell": false, + "toc_section_display": "block", + "toc_window_display": false, + "widenNotebook": false + } + }, + "nbformat": 4, + "nbformat_minor": 2 +}